Upload 19 files
Browse files- .gitattributes +1 -0
- 1_Pooling/config.json +10 -0
- added_tokens.json +28 -0
- args.json +385 -0
- chat_template.jinja +85 -0
- config.json +60 -0
- config_sentence_transformers.json +8 -0
- generation_config.json +10 -0
- latest +1 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- modules.json +20 -0
- scheduler.pt +3 -0
- special_tokens_map.json +31 -0
- tokenizer.json +3 -0
- tokenizer_config.json +239 -0
- trainer_state.json +2759 -0
- training_args.bin +3 -0
- vocab.json +0 -0
- zero_to_fp32.py +760 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
1_Pooling/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"word_embedding_dimension": 1024,
|
| 3 |
+
"pooling_mode_cls_token": false,
|
| 4 |
+
"pooling_mode_mean_tokens": false,
|
| 5 |
+
"pooling_mode_max_tokens": false,
|
| 6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
+
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
+
"pooling_mode_lasttoken": true,
|
| 9 |
+
"include_prompt": true
|
| 10 |
+
}
|
added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
args.json
ADDED
|
@@ -0,0 +1,385 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"output_dir": "/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025",
|
| 3 |
+
"overwrite_output_dir": false,
|
| 4 |
+
"do_train": false,
|
| 5 |
+
"do_eval": false,
|
| 6 |
+
"do_predict": false,
|
| 7 |
+
"eval_strategy": "steps",
|
| 8 |
+
"prediction_loss_only": false,
|
| 9 |
+
"per_device_train_batch_size": 4,
|
| 10 |
+
"per_device_eval_batch_size": 4,
|
| 11 |
+
"per_gpu_train_batch_size": null,
|
| 12 |
+
"per_gpu_eval_batch_size": null,
|
| 13 |
+
"gradient_accumulation_steps": 4,
|
| 14 |
+
"eval_accumulation_steps": null,
|
| 15 |
+
"eval_delay": 0,
|
| 16 |
+
"torch_empty_cache_steps": null,
|
| 17 |
+
"learning_rate": 6e-06,
|
| 18 |
+
"weight_decay": 0.1,
|
| 19 |
+
"adam_beta1": 0.9,
|
| 20 |
+
"adam_beta2": 0.95,
|
| 21 |
+
"adam_epsilon": 1e-08,
|
| 22 |
+
"max_grad_norm": 1.0,
|
| 23 |
+
"num_train_epochs": 5.0,
|
| 24 |
+
"max_steps": -1,
|
| 25 |
+
"lr_scheduler_type": "cosine",
|
| 26 |
+
"lr_scheduler_kwargs": null,
|
| 27 |
+
"warmup_ratio": 0.0,
|
| 28 |
+
"warmup_steps": 0,
|
| 29 |
+
"log_level": "passive",
|
| 30 |
+
"log_level_replica": "warning",
|
| 31 |
+
"log_on_each_node": true,
|
| 32 |
+
"logging_dir": "/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025/runs",
|
| 33 |
+
"logging_strategy": "steps",
|
| 34 |
+
"logging_first_step": true,
|
| 35 |
+
"logging_steps": 5,
|
| 36 |
+
"logging_nan_inf_filter": true,
|
| 37 |
+
"save_strategy": "steps",
|
| 38 |
+
"save_steps": 100.0,
|
| 39 |
+
"save_total_limit": null,
|
| 40 |
+
"save_safetensors": true,
|
| 41 |
+
"save_on_each_node": false,
|
| 42 |
+
"save_only_model": false,
|
| 43 |
+
"restore_callback_states_from_checkpoint": false,
|
| 44 |
+
"no_cuda": false,
|
| 45 |
+
"use_cpu": false,
|
| 46 |
+
"use_mps_device": false,
|
| 47 |
+
"seed": 42,
|
| 48 |
+
"data_seed": 42,
|
| 49 |
+
"jit_mode_eval": false,
|
| 50 |
+
"bf16": true,
|
| 51 |
+
"fp16": false,
|
| 52 |
+
"fp16_opt_level": "O1",
|
| 53 |
+
"half_precision_backend": "auto",
|
| 54 |
+
"bf16_full_eval": false,
|
| 55 |
+
"fp16_full_eval": false,
|
| 56 |
+
"tf32": null,
|
| 57 |
+
"local_rank": 0,
|
| 58 |
+
"ddp_backend": null,
|
| 59 |
+
"tpu_num_cores": null,
|
| 60 |
+
"tpu_metrics_debug": false,
|
| 61 |
+
"debug": null,
|
| 62 |
+
"dataloader_drop_last": true,
|
| 63 |
+
"eval_steps": 100.0,
|
| 64 |
+
"dataloader_num_workers": null,
|
| 65 |
+
"dataloader_prefetch_factor": null,
|
| 66 |
+
"past_index": -1,
|
| 67 |
+
"run_name": "/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025",
|
| 68 |
+
"disable_tqdm": null,
|
| 69 |
+
"remove_unused_columns": true,
|
| 70 |
+
"label_names": [
|
| 71 |
+
"labels"
|
| 72 |
+
],
|
| 73 |
+
"load_best_model_at_end": false,
|
| 74 |
+
"metric_for_best_model": "loss",
|
| 75 |
+
"greater_is_better": false,
|
| 76 |
+
"ignore_data_skip": false,
|
| 77 |
+
"fsdp": null,
|
| 78 |
+
"fsdp_min_num_params": 0,
|
| 79 |
+
"fsdp_config": null,
|
| 80 |
+
"fsdp_transformer_layer_cls_to_wrap": null,
|
| 81 |
+
"accelerator_config": {
|
| 82 |
+
"dispatch_batches": false
|
| 83 |
+
},
|
| 84 |
+
"parallelism_config": null,
|
| 85 |
+
"deepspeed": {
|
| 86 |
+
"fp16": {
|
| 87 |
+
"enabled": "auto",
|
| 88 |
+
"loss_scale": 0,
|
| 89 |
+
"loss_scale_window": 1000,
|
| 90 |
+
"initial_scale_power": 16,
|
| 91 |
+
"hysteresis": 2,
|
| 92 |
+
"min_loss_scale": 1
|
| 93 |
+
},
|
| 94 |
+
"bf16": {
|
| 95 |
+
"enabled": "auto"
|
| 96 |
+
},
|
| 97 |
+
"zero_optimization": {
|
| 98 |
+
"stage": 3,
|
| 99 |
+
"offload_optimizer": {
|
| 100 |
+
"device": "none",
|
| 101 |
+
"pin_memory": true
|
| 102 |
+
},
|
| 103 |
+
"offload_param": {
|
| 104 |
+
"device": "none",
|
| 105 |
+
"pin_memory": true
|
| 106 |
+
},
|
| 107 |
+
"overlap_comm": false,
|
| 108 |
+
"contiguous_gradients": true,
|
| 109 |
+
"sub_group_size": 1000000000.0,
|
| 110 |
+
"reduce_bucket_size": "auto",
|
| 111 |
+
"zero_quantized_weights": false,
|
| 112 |
+
"zero_quantized_gradients": false,
|
| 113 |
+
"stage3_prefetch_bucket_size": "auto",
|
| 114 |
+
"stage3_param_persistence_threshold": "auto",
|
| 115 |
+
"stage3_max_live_parameters": 1000000000.0,
|
| 116 |
+
"stage3_max_reuse_distance": 1000000000.0,
|
| 117 |
+
"stage3_gather_16bit_weights_on_model_save": true
|
| 118 |
+
},
|
| 119 |
+
"gradient_accumulation_steps": "auto",
|
| 120 |
+
"gradient_clipping": "auto",
|
| 121 |
+
"steps_per_print": 2000,
|
| 122 |
+
"train_batch_size": "auto",
|
| 123 |
+
"train_micro_batch_size_per_gpu": "auto",
|
| 124 |
+
"wall_clock_breakdown": false
|
| 125 |
+
},
|
| 126 |
+
"label_smoothing_factor": 0.0,
|
| 127 |
+
"optim": "adamw_torch",
|
| 128 |
+
"optim_args": null,
|
| 129 |
+
"adafactor": false,
|
| 130 |
+
"group_by_length": false,
|
| 131 |
+
"length_column_name": "length",
|
| 132 |
+
"report_to": [
|
| 133 |
+
"tensorboard"
|
| 134 |
+
],
|
| 135 |
+
"project": "huggingface",
|
| 136 |
+
"trackio_space_id": "trackio",
|
| 137 |
+
"ddp_find_unused_parameters": null,
|
| 138 |
+
"ddp_bucket_cap_mb": null,
|
| 139 |
+
"ddp_broadcast_buffers": null,
|
| 140 |
+
"dataloader_pin_memory": true,
|
| 141 |
+
"dataloader_persistent_workers": false,
|
| 142 |
+
"skip_memory_metrics": true,
|
| 143 |
+
"use_legacy_prediction_loop": false,
|
| 144 |
+
"push_to_hub": false,
|
| 145 |
+
"resume_from_checkpoint": null,
|
| 146 |
+
"hub_model_id": null,
|
| 147 |
+
"hub_strategy": "every_save",
|
| 148 |
+
"hub_token": null,
|
| 149 |
+
"hub_private_repo": null,
|
| 150 |
+
"hub_always_push": false,
|
| 151 |
+
"hub_revision": null,
|
| 152 |
+
"gradient_checkpointing": true,
|
| 153 |
+
"gradient_checkpointing_kwargs": null,
|
| 154 |
+
"include_inputs_for_metrics": false,
|
| 155 |
+
"include_for_metrics": [],
|
| 156 |
+
"eval_do_concat_batches": true,
|
| 157 |
+
"fp16_backend": "auto",
|
| 158 |
+
"push_to_hub_model_id": null,
|
| 159 |
+
"push_to_hub_organization": null,
|
| 160 |
+
"push_to_hub_token": null,
|
| 161 |
+
"mp_parameters": "",
|
| 162 |
+
"auto_find_batch_size": false,
|
| 163 |
+
"full_determinism": false,
|
| 164 |
+
"torchdynamo": null,
|
| 165 |
+
"ray_scope": "last",
|
| 166 |
+
"ddp_timeout": 18000000,
|
| 167 |
+
"torch_compile": false,
|
| 168 |
+
"torch_compile_backend": null,
|
| 169 |
+
"torch_compile_mode": null,
|
| 170 |
+
"include_tokens_per_second": false,
|
| 171 |
+
"include_num_input_tokens_seen": false,
|
| 172 |
+
"neftune_noise_alpha": null,
|
| 173 |
+
"optim_target_modules": null,
|
| 174 |
+
"batch_eval_metrics": false,
|
| 175 |
+
"eval_on_start": false,
|
| 176 |
+
"use_liger_kernel": false,
|
| 177 |
+
"liger_kernel_config": null,
|
| 178 |
+
"eval_use_gather_object": false,
|
| 179 |
+
"average_tokens_across_devices": true,
|
| 180 |
+
"sortish_sampler": false,
|
| 181 |
+
"predict_with_generate": false,
|
| 182 |
+
"generation_max_length": null,
|
| 183 |
+
"generation_num_beams": null,
|
| 184 |
+
"generation_config": null,
|
| 185 |
+
"tuner_backend": "peft",
|
| 186 |
+
"vit_gradient_checkpointing": null,
|
| 187 |
+
"router_aux_loss_coef": 0.0,
|
| 188 |
+
"enable_dft_loss": false,
|
| 189 |
+
"enable_channel_loss": false,
|
| 190 |
+
"check_model": true,
|
| 191 |
+
"acc_strategy": "token",
|
| 192 |
+
"train_dataloader_shuffle": true,
|
| 193 |
+
"max_epochs": null,
|
| 194 |
+
"aligner_lr": null,
|
| 195 |
+
"vit_lr": null,
|
| 196 |
+
"use_logits_to_keep": null,
|
| 197 |
+
"ds3_gather_for_generation": true,
|
| 198 |
+
"resume_only_model": false,
|
| 199 |
+
"optimizer": null,
|
| 200 |
+
"loss_type": "infonce",
|
| 201 |
+
"metric": null,
|
| 202 |
+
"eval_use_evalscope": false,
|
| 203 |
+
"eval_dataset": [],
|
| 204 |
+
"eval_dataset_args": null,
|
| 205 |
+
"eval_limit": null,
|
| 206 |
+
"eval_generation_config": null,
|
| 207 |
+
"extra_eval_args": null,
|
| 208 |
+
"use_flash_ckpt": false,
|
| 209 |
+
"model": "/mnt/beegfs3/liying/zhangfanhao/Qwen3-Embedding-0.6B",
|
| 210 |
+
"model_type": "qwen3_emb",
|
| 211 |
+
"model_revision": null,
|
| 212 |
+
"task_type": "embedding",
|
| 213 |
+
"torch_dtype": "bfloat16",
|
| 214 |
+
"attn_impl": null,
|
| 215 |
+
"new_special_tokens": [],
|
| 216 |
+
"num_labels": null,
|
| 217 |
+
"problem_type": null,
|
| 218 |
+
"rope_scaling": null,
|
| 219 |
+
"device_map": null,
|
| 220 |
+
"max_memory": {},
|
| 221 |
+
"max_model_len": null,
|
| 222 |
+
"local_repo_path": null,
|
| 223 |
+
"init_strategy": null,
|
| 224 |
+
"template": "qwen3_emb",
|
| 225 |
+
"system": null,
|
| 226 |
+
"max_length": 2048,
|
| 227 |
+
"truncation_strategy": "left",
|
| 228 |
+
"max_pixels": null,
|
| 229 |
+
"agent_template": null,
|
| 230 |
+
"norm_bbox": null,
|
| 231 |
+
"use_chat_template": true,
|
| 232 |
+
"padding_free": false,
|
| 233 |
+
"padding_side": "right",
|
| 234 |
+
"loss_scale": "default",
|
| 235 |
+
"sequence_parallel_size": 1,
|
| 236 |
+
"response_prefix": null,
|
| 237 |
+
"template_backend": "swift",
|
| 238 |
+
"dataset": [
|
| 239 |
+
"/mnt/beegfs3/liying/zhangfanhao/testdata_final_all_1126.jsonl"
|
| 240 |
+
],
|
| 241 |
+
"val_dataset": [],
|
| 242 |
+
"split_dataset_ratio": 0.05,
|
| 243 |
+
"dataset_num_proc": 32,
|
| 244 |
+
"load_from_cache_file": false,
|
| 245 |
+
"dataset_shuffle": true,
|
| 246 |
+
"val_dataset_shuffle": false,
|
| 247 |
+
"streaming": false,
|
| 248 |
+
"interleave_prob": null,
|
| 249 |
+
"stopping_strategy": "first_exhausted",
|
| 250 |
+
"shuffle_buffer_size": 1000,
|
| 251 |
+
"download_mode": "reuse_dataset_if_exists",
|
| 252 |
+
"columns": {},
|
| 253 |
+
"strict": false,
|
| 254 |
+
"model_name": null,
|
| 255 |
+
"model_author": null,
|
| 256 |
+
"custom_dataset_info": [],
|
| 257 |
+
"quant_method": null,
|
| 258 |
+
"quant_bits": null,
|
| 259 |
+
"hqq_axis": null,
|
| 260 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 261 |
+
"bnb_4bit_quant_type": "nf4",
|
| 262 |
+
"bnb_4bit_use_double_quant": true,
|
| 263 |
+
"bnb_4bit_quant_storage": null,
|
| 264 |
+
"max_new_tokens": 64,
|
| 265 |
+
"temperature": 0.0,
|
| 266 |
+
"top_k": null,
|
| 267 |
+
"top_p": null,
|
| 268 |
+
"repetition_penalty": null,
|
| 269 |
+
"num_beams": 1,
|
| 270 |
+
"stream": false,
|
| 271 |
+
"stop_words": [],
|
| 272 |
+
"logprobs": false,
|
| 273 |
+
"top_logprobs": null,
|
| 274 |
+
"ckpt_dir": null,
|
| 275 |
+
"lora_modules": [],
|
| 276 |
+
"train_type": "full",
|
| 277 |
+
"adapters": [],
|
| 278 |
+
"external_plugins": [],
|
| 279 |
+
"model_kwargs": {},
|
| 280 |
+
"load_args": false,
|
| 281 |
+
"load_data_args": false,
|
| 282 |
+
"packing": false,
|
| 283 |
+
"packing_length": null,
|
| 284 |
+
"lazy_tokenize": false,
|
| 285 |
+
"cached_dataset": [],
|
| 286 |
+
"custom_register_path": [],
|
| 287 |
+
"use_hf": false,
|
| 288 |
+
"ignore_args_error": false,
|
| 289 |
+
"use_swift_lora": false,
|
| 290 |
+
"freeze_parameters": [],
|
| 291 |
+
"freeze_parameters_regex": null,
|
| 292 |
+
"freeze_parameters_ratio": 0.0,
|
| 293 |
+
"trainable_parameters": [],
|
| 294 |
+
"trainable_parameters_regex": null,
|
| 295 |
+
"freeze_llm": false,
|
| 296 |
+
"freeze_vit": true,
|
| 297 |
+
"freeze_aligner": true,
|
| 298 |
+
"target_modules": [
|
| 299 |
+
"all-linear"
|
| 300 |
+
],
|
| 301 |
+
"target_regex": null,
|
| 302 |
+
"target_parameters": null,
|
| 303 |
+
"modules_to_save": [],
|
| 304 |
+
"lora_rank": 8,
|
| 305 |
+
"lora_alpha": 32,
|
| 306 |
+
"lora_dropout": 0.05,
|
| 307 |
+
"lora_bias": "none",
|
| 308 |
+
"lora_dtype": null,
|
| 309 |
+
"lorap_lr_ratio": null,
|
| 310 |
+
"use_rslora": false,
|
| 311 |
+
"use_dora": false,
|
| 312 |
+
"lora_ga_batch_size": 2,
|
| 313 |
+
"lora_ga_iters": 2,
|
| 314 |
+
"lora_ga_max_length": 1024,
|
| 315 |
+
"lora_ga_direction": "ArB2r",
|
| 316 |
+
"lora_ga_scale": "stable",
|
| 317 |
+
"lora_ga_stable_gamma": 16,
|
| 318 |
+
"init_weights": true,
|
| 319 |
+
"fourier_n_frequency": 2000,
|
| 320 |
+
"fourier_scaling": 300.0,
|
| 321 |
+
"boft_block_size": 4,
|
| 322 |
+
"boft_block_num": 0,
|
| 323 |
+
"boft_n_butterfly_factor": 1,
|
| 324 |
+
"boft_dropout": 0.0,
|
| 325 |
+
"vera_rank": 256,
|
| 326 |
+
"vera_projection_prng_key": 0,
|
| 327 |
+
"vera_dropout": 0.0,
|
| 328 |
+
"vera_d_initial": 0.1,
|
| 329 |
+
"adapter_act": "gelu",
|
| 330 |
+
"adapter_length": 128,
|
| 331 |
+
"use_galore": false,
|
| 332 |
+
"galore_target_modules": null,
|
| 333 |
+
"galore_rank": 128,
|
| 334 |
+
"galore_update_proj_gap": 50,
|
| 335 |
+
"galore_scale": 1.0,
|
| 336 |
+
"galore_proj_type": "std",
|
| 337 |
+
"galore_optim_per_parameter": false,
|
| 338 |
+
"galore_with_embedding": false,
|
| 339 |
+
"galore_quantization": false,
|
| 340 |
+
"galore_proj_quant": false,
|
| 341 |
+
"galore_proj_bits": 4,
|
| 342 |
+
"galore_proj_group_size": 256,
|
| 343 |
+
"galore_cos_threshold": 0.4,
|
| 344 |
+
"galore_gamma_proj": 2,
|
| 345 |
+
"galore_queue_size": 5,
|
| 346 |
+
"adalora_target_r": 8,
|
| 347 |
+
"adalora_init_r": 12,
|
| 348 |
+
"adalora_tinit": 0,
|
| 349 |
+
"adalora_tfinal": 0,
|
| 350 |
+
"adalora_deltaT": 1,
|
| 351 |
+
"adalora_beta1": 0.85,
|
| 352 |
+
"adalora_beta2": 0.85,
|
| 353 |
+
"adalora_orth_reg_weight": 0.5,
|
| 354 |
+
"llamapro_num_new_blocks": 4,
|
| 355 |
+
"llamapro_num_groups": null,
|
| 356 |
+
"lisa_activated_layers": 0,
|
| 357 |
+
"lisa_step_interval": 20,
|
| 358 |
+
"reft_layer_key": null,
|
| 359 |
+
"reft_layers": null,
|
| 360 |
+
"reft_rank": 4,
|
| 361 |
+
"reft_intervention_type": "LoreftIntervention",
|
| 362 |
+
"reft_args": null,
|
| 363 |
+
"swanlab_token": null,
|
| 364 |
+
"swanlab_project": null,
|
| 365 |
+
"swanlab_workspace": null,
|
| 366 |
+
"swanlab_exp_name": null,
|
| 367 |
+
"swanlab_lark_webhook_url": null,
|
| 368 |
+
"swanlab_lark_secret": null,
|
| 369 |
+
"swanlab_mode": "cloud",
|
| 370 |
+
"add_version": true,
|
| 371 |
+
"create_checkpoint_symlink": false,
|
| 372 |
+
"zero_hpz_partition_size": null,
|
| 373 |
+
"deepspeed_autotp_size": null,
|
| 374 |
+
"early_stop_interval": null,
|
| 375 |
+
"rank": 0,
|
| 376 |
+
"global_world_size": 16,
|
| 377 |
+
"local_world_size": 8,
|
| 378 |
+
"model_suffix": "Qwen3-Embedding-0.6B",
|
| 379 |
+
"model_info": "ModelInfo(model_type='qwen3_emb', model_dir='/mnt/beegfs3/liying/zhangfanhao/Qwen3-Embedding-0.6B', torch_dtype=torch.bfloat16, max_model_len=32768, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, config=None, task_type='embedding', num_labels=None)",
|
| 380 |
+
"model_meta": "ModelMeta(model_type='qwen3_emb', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-Embedding-0.6B', hf_model_id='Qwen/Qwen3-Embedding-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-Embedding-4B', hf_model_id='Qwen/Qwen3-Embedding-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-Embedding-8B', hf_model_id='Qwen/Qwen3-Embedding-8B', model_path=None, ms_revision=None, hf_revision=None)], ignore_patterns=None, requires=None, tags=[])], template='qwen3_emb', get_function=<function get_model_tokenizer_with_flash_attn at 0x7fa2e318c4a0>, model_arch=None, architectures=['Qwen3ForCausalLM'], additional_saved_files=['config_sentence_transformers.json', '1_Pooling', 'modules.json'], torch_dtype=None, is_multimodal=False, is_reward=False, task_type=None, ignore_patterns=None, requires=[], tags=[])",
|
| 381 |
+
"model_dir": "/mnt/beegfs3/liying/zhangfanhao/Qwen3-Embedding-0.6B",
|
| 382 |
+
"hub": "<class 'swift.hub.hub.MSHub'>",
|
| 383 |
+
"evaluation_strategy": "steps",
|
| 384 |
+
"training_args": "TrainingArguments(output_dir='/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=4, per_device_eval_batch_size=4, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=4, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=6e-06, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=5.0, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_ratio=0.0, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=5, logging_nan_inf_filter=True, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=100, save_total_limit=None, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=42, jit_mode_eval=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=True, eval_steps=100, dataloader_num_workers=1, dataloader_prefetch_factor=10, past_index=-1, run_name='/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025', disable_tqdm=False, remove_unused_columns=False, label_names=['labels'], load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), parallelism_config=None, deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 3, 'offload_optimizer': {'device': 'none', 'pin_memory': True}, 'offload_param': {'device': 'none', 'pin_memory': True}, 'overlap_comm': False, 'contiguous_gradients': True, 'sub_group_size': 1000000000.0, 'reduce_bucket_size': 'auto', 'zero_quantized_weights': False, 'zero_quantized_gradients': False, 'stage3_prefetch_bucket_size': 'auto', 'stage3_param_persistence_threshold': 'auto', 'stage3_max_live_parameters': 1000000000.0, 'stage3_max_reuse_distance': 1000000000.0, 'stage3_gather_16bit_weights_on_model_save': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], project='huggingface', trackio_space_id='trackio', ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=None, hub_always_push=False, hub_revision=None, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, include_for_metrics=[], eval_do_concat_batches=True, fp16_backend='auto', push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=18000000, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, include_tokens_per_second=None, include_num_input_tokens_seen=None, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, use_liger_kernel=False, liger_kernel_config=None, eval_use_gather_object=False, average_tokens_across_devices=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, check_model=True, acc_strategy='token', train_dataloader_shuffle=True, max_epochs=None, aligner_lr=None, vit_lr=None, use_logits_to_keep=None, ds3_gather_for_generation=True, resume_only_model=False, optimizer=None, loss_type='infonce', metric=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, use_flash_ckpt=False, sft_alpha=0, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, train_type='full', local_repo_path=None, galore_config=None, padding_side='right', padding_free=False, task_type='embedding')"
|
| 385 |
+
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 27 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 28 |
+
{%- elif message.role == "assistant" %}
|
| 29 |
+
{%- set content = message.content %}
|
| 30 |
+
{%- set reasoning_content = '' %}
|
| 31 |
+
{%- if message.reasoning_content is defined and message.reasoning_content is not none %}
|
| 32 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 33 |
+
{%- else %}
|
| 34 |
+
{%- if '</think>' in message.content %}
|
| 35 |
+
{%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
|
| 36 |
+
{%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 37 |
+
{%- endif %}
|
| 38 |
+
{%- endif %}
|
| 39 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 40 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 41 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 42 |
+
{%- else %}
|
| 43 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- else %}
|
| 46 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 47 |
+
{%- endif %}
|
| 48 |
+
{%- if message.tool_calls %}
|
| 49 |
+
{%- for tool_call in message.tool_calls %}
|
| 50 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 51 |
+
{{- '\n' }}
|
| 52 |
+
{%- endif %}
|
| 53 |
+
{%- if tool_call.function %}
|
| 54 |
+
{%- set tool_call = tool_call.function %}
|
| 55 |
+
{%- endif %}
|
| 56 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 57 |
+
{{- tool_call.name }}
|
| 58 |
+
{{- '", "arguments": ' }}
|
| 59 |
+
{%- if tool_call.arguments is string %}
|
| 60 |
+
{{- tool_call.arguments }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{{- tool_call.arguments | tojson }}
|
| 63 |
+
{%- endif %}
|
| 64 |
+
{{- '}\n</tool_call>' }}
|
| 65 |
+
{%- endfor %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{{- '<|im_end|>\n' }}
|
| 68 |
+
{%- elif message.role == "tool" %}
|
| 69 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 70 |
+
{{- '<|im_start|>user' }}
|
| 71 |
+
{%- endif %}
|
| 72 |
+
{{- '\n<tool_response>\n' }}
|
| 73 |
+
{{- message.content }}
|
| 74 |
+
{{- '\n</tool_response>' }}
|
| 75 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 76 |
+
{{- '<|im_end|>\n' }}
|
| 77 |
+
{%- endif %}
|
| 78 |
+
{%- endif %}
|
| 79 |
+
{%- endfor %}
|
| 80 |
+
{%- if add_generation_prompt %}
|
| 81 |
+
{{- '<|im_start|>assistant\n' }}
|
| 82 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 83 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 84 |
+
{%- endif %}
|
| 85 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eos_token_id": 151645,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_types": [
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention"
|
| 43 |
+
],
|
| 44 |
+
"max_position_embeddings": 32768,
|
| 45 |
+
"max_window_layers": 28,
|
| 46 |
+
"model_type": "qwen3",
|
| 47 |
+
"num_attention_heads": 16,
|
| 48 |
+
"num_hidden_layers": 28,
|
| 49 |
+
"num_key_value_heads": 8,
|
| 50 |
+
"pad_token_id": 151643,
|
| 51 |
+
"rms_norm_eps": 1e-06,
|
| 52 |
+
"rope_scaling": null,
|
| 53 |
+
"rope_theta": 1000000,
|
| 54 |
+
"sliding_window": null,
|
| 55 |
+
"tie_word_embeddings": true,
|
| 56 |
+
"transformers_version": "4.57.1",
|
| 57 |
+
"use_cache": false,
|
| 58 |
+
"use_sliding_window": false,
|
| 59 |
+
"vocab_size": 151669
|
| 60 |
+
}
|
config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"prompts": {
|
| 3 |
+
"query": "Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery:",
|
| 4 |
+
"document": ""
|
| 5 |
+
},
|
| 6 |
+
"default_prompt_name": null,
|
| 7 |
+
"similarity_fn_name": "cosine"
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token_id": [
|
| 3 |
+
151645,
|
| 4 |
+
151643,
|
| 5 |
+
151645
|
| 6 |
+
],
|
| 7 |
+
"max_new_tokens": 2048,
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"transformers_version": "4.57.1"
|
| 10 |
+
}
|
latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step1800
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7b814857c0036884b7e6315489fd79ac32f5c755c6ce719e45c0c08d3209df0
|
| 3 |
+
size 1191588280
|
modules.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"idx": 0,
|
| 4 |
+
"name": "0",
|
| 5 |
+
"path": "",
|
| 6 |
+
"type": "sentence_transformers.models.Transformer"
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"idx": 1,
|
| 10 |
+
"name": "1",
|
| 11 |
+
"path": "1_Pooling",
|
| 12 |
+
"type": "sentence_transformers.models.Pooling"
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"idx": 2,
|
| 16 |
+
"name": "2",
|
| 17 |
+
"path": "2_Normalize",
|
| 18 |
+
"type": "sentence_transformers.models.Normalize"
|
| 19 |
+
}
|
| 20 |
+
]
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3df3e7f8a6cd70268712b0cc8a07baf070597f77383ff0df8999c53214c03ba8
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|im_end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": {
|
| 25 |
+
"content": "<|endoftext|>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
}
|
| 31 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:def76fb086971c7867b829c23a26261e38d9d74e02139253b38aeb9df8b4b50a
|
| 3 |
+
size 11423705
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,239 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"clean_up_tokenization_spaces": false,
|
| 231 |
+
"eos_token": "<|im_end|>",
|
| 232 |
+
"errors": "replace",
|
| 233 |
+
"extra_special_tokens": {},
|
| 234 |
+
"model_max_length": 131072,
|
| 235 |
+
"pad_token": "<|endoftext|>",
|
| 236 |
+
"split_special_tokens": false,
|
| 237 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 238 |
+
"unk_token": null
|
| 239 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2759 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1800,
|
| 3 |
+
"best_metric": 0.00229549,
|
| 4 |
+
"best_model_checkpoint": "/mnt/beegfs3/liying/zhangfanhao/output1125/v1-20251125-231025/checkpoint-1800",
|
| 5 |
+
"epoch": 3.0354280894137493,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.001687051876845213,
|
| 14 |
+
"grad_norm": 0.5938383277366354,
|
| 15 |
+
"learning_rate": 5.999998316002012e-06,
|
| 16 |
+
"loss": 0.380859375,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008435259384226065,
|
| 21 |
+
"grad_norm": 0.432332139447319,
|
| 22 |
+
"learning_rate": 5.999957900144816e-06,
|
| 23 |
+
"loss": 0.3326416015625,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01687051876845213,
|
| 28 |
+
"grad_norm": 0.24245712798777588,
|
| 29 |
+
"learning_rate": 5.99983160176086e-06,
|
| 30 |
+
"loss": 0.2187744140625,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.025305778152678194,
|
| 35 |
+
"grad_norm": 0.14967602144842607,
|
| 36 |
+
"learning_rate": 5.999621108392896e-06,
|
| 37 |
+
"loss": 0.1771240234375,
|
| 38 |
+
"step": 15
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03374103753690426,
|
| 42 |
+
"grad_norm": 0.11965916268612647,
|
| 43 |
+
"learning_rate": 5.9993264259487505e-06,
|
| 44 |
+
"loss": 0.14423828125,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04217629692113033,
|
| 49 |
+
"grad_norm": 0.09039362542123534,
|
| 50 |
+
"learning_rate": 5.998947562699149e-06,
|
| 51 |
+
"loss": 0.1184326171875,
|
| 52 |
+
"step": 25
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05061155630535639,
|
| 56 |
+
"grad_norm": 0.0761794885482189,
|
| 57 |
+
"learning_rate": 5.998484529277483e-06,
|
| 58 |
+
"loss": 0.108642578125,
|
| 59 |
+
"step": 30
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.059046815689582456,
|
| 63 |
+
"grad_norm": 0.07436752367684027,
|
| 64 |
+
"learning_rate": 5.997937338679513e-06,
|
| 65 |
+
"loss": 0.09638671875,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06748207507380852,
|
| 70 |
+
"grad_norm": 0.06054003854062884,
|
| 71 |
+
"learning_rate": 5.997306006263003e-06,
|
| 72 |
+
"loss": 0.1025146484375,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07591733445803459,
|
| 77 |
+
"grad_norm": 0.05715450839425674,
|
| 78 |
+
"learning_rate": 5.996590549747288e-06,
|
| 79 |
+
"loss": 0.0909912109375,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08435259384226065,
|
| 84 |
+
"grad_norm": 0.0562159873926997,
|
| 85 |
+
"learning_rate": 5.995790989212777e-06,
|
| 86 |
+
"loss": 0.0900390625,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09278785322648671,
|
| 91 |
+
"grad_norm": 0.054733644360014155,
|
| 92 |
+
"learning_rate": 5.994907347100393e-06,
|
| 93 |
+
"loss": 0.08599853515625,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10122311261071278,
|
| 98 |
+
"grad_norm": 0.04945430208391664,
|
| 99 |
+
"learning_rate": 5.99393964821094e-06,
|
| 100 |
+
"loss": 0.08861083984375,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10965837199493884,
|
| 105 |
+
"grad_norm": 0.060375343186170424,
|
| 106 |
+
"learning_rate": 5.992887919704406e-06,
|
| 107 |
+
"loss": 0.08037109375,
|
| 108 |
+
"step": 65
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11809363137916491,
|
| 112 |
+
"grad_norm": 0.05113371142226039,
|
| 113 |
+
"learning_rate": 5.991752191099203e-06,
|
| 114 |
+
"loss": 0.07867431640625,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12652889076339097,
|
| 119 |
+
"grad_norm": 0.060073186423122656,
|
| 120 |
+
"learning_rate": 5.990532494271337e-06,
|
| 121 |
+
"loss": 0.07816162109375,
|
| 122 |
+
"step": 75
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13496415014761703,
|
| 126 |
+
"grad_norm": 0.058832653609599356,
|
| 127 |
+
"learning_rate": 5.989228863453515e-06,
|
| 128 |
+
"loss": 0.08001708984375,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1433994095318431,
|
| 133 |
+
"grad_norm": 0.06587176624760811,
|
| 134 |
+
"learning_rate": 5.987841335234184e-06,
|
| 135 |
+
"loss": 0.074359130859375,
|
| 136 |
+
"step": 85
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.15183466891606917,
|
| 140 |
+
"grad_norm": 0.06491166432460505,
|
| 141 |
+
"learning_rate": 5.9863699485565e-06,
|
| 142 |
+
"loss": 0.0674072265625,
|
| 143 |
+
"step": 90
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16026992830029524,
|
| 147 |
+
"grad_norm": 0.056198676389375694,
|
| 148 |
+
"learning_rate": 5.984814744717241e-06,
|
| 149 |
+
"loss": 0.0659912109375,
|
| 150 |
+
"step": 95
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.1687051876845213,
|
| 154 |
+
"grad_norm": 0.0673764252680421,
|
| 155 |
+
"learning_rate": 5.983175767365646e-06,
|
| 156 |
+
"loss": 0.063623046875,
|
| 157 |
+
"step": 100
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1687051876845213,
|
| 161 |
+
"eval_loss": 0.0162808820605278,
|
| 162 |
+
"eval_margin": -0.020074697267714766,
|
| 163 |
+
"eval_mean_neg": 0.6548054814338684,
|
| 164 |
+
"eval_mean_pos": 0.8441178202629089,
|
| 165 |
+
"eval_runtime": 367.938,
|
| 166 |
+
"eval_samples_per_second": 21.713,
|
| 167 |
+
"eval_steps_per_second": 0.34,
|
| 168 |
+
"step": 100
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.17714044706874738,
|
| 172 |
+
"grad_norm": 0.06055978762872105,
|
| 173 |
+
"learning_rate": 5.981453062502185e-06,
|
| 174 |
+
"loss": 0.060498046875,
|
| 175 |
+
"step": 105
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.18557570645297342,
|
| 179 |
+
"grad_norm": 0.06213709235940642,
|
| 180 |
+
"learning_rate": 5.979646678477277e-06,
|
| 181 |
+
"loss": 0.056640625,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.19401096583719948,
|
| 186 |
+
"grad_norm": 0.0659729457413995,
|
| 187 |
+
"learning_rate": 5.977756665989925e-06,
|
| 188 |
+
"loss": 0.05919189453125,
|
| 189 |
+
"step": 115
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.20244622522142555,
|
| 193 |
+
"grad_norm": 0.05897713608413389,
|
| 194 |
+
"learning_rate": 5.9757830780862985e-06,
|
| 195 |
+
"loss": 0.0628662109375,
|
| 196 |
+
"step": 120
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.21088148460565162,
|
| 200 |
+
"grad_norm": 0.05642517065149083,
|
| 201 |
+
"learning_rate": 5.973725970158239e-06,
|
| 202 |
+
"loss": 0.05245361328125,
|
| 203 |
+
"step": 125
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.2193167439898777,
|
| 207 |
+
"grad_norm": 0.06579611075607034,
|
| 208 |
+
"learning_rate": 5.9715853999417115e-06,
|
| 209 |
+
"loss": 0.05848388671875,
|
| 210 |
+
"step": 130
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.22775200337410376,
|
| 214 |
+
"grad_norm": 0.07448489445734133,
|
| 215 |
+
"learning_rate": 5.969361427515179e-06,
|
| 216 |
+
"loss": 0.0573974609375,
|
| 217 |
+
"step": 135
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.23618726275832982,
|
| 221 |
+
"grad_norm": 0.05975086799089143,
|
| 222 |
+
"learning_rate": 5.9670541152979215e-06,
|
| 223 |
+
"loss": 0.05091552734375,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.2446225221425559,
|
| 228 |
+
"grad_norm": 0.06504159374670346,
|
| 229 |
+
"learning_rate": 5.964663528048276e-06,
|
| 230 |
+
"loss": 0.047943115234375,
|
| 231 |
+
"step": 145
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.25305778152678193,
|
| 235 |
+
"grad_norm": 0.060584307441235295,
|
| 236 |
+
"learning_rate": 5.96218973286183e-06,
|
| 237 |
+
"loss": 0.0493896484375,
|
| 238 |
+
"step": 150
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.261493040911008,
|
| 242 |
+
"grad_norm": 0.06234398910972033,
|
| 243 |
+
"learning_rate": 5.959632799169529e-06,
|
| 244 |
+
"loss": 0.04854736328125,
|
| 245 |
+
"step": 155
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.26992830029523407,
|
| 249 |
+
"grad_norm": 0.07326440644425879,
|
| 250 |
+
"learning_rate": 5.9569927987357305e-06,
|
| 251 |
+
"loss": 0.0443359375,
|
| 252 |
+
"step": 160
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.27836355967946014,
|
| 256 |
+
"grad_norm": 0.059873291821439245,
|
| 257 |
+
"learning_rate": 5.954269805656194e-06,
|
| 258 |
+
"loss": 0.04698486328125,
|
| 259 |
+
"step": 165
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.2867988190636862,
|
| 263 |
+
"grad_norm": 0.07214278446872342,
|
| 264 |
+
"learning_rate": 5.951463896355993e-06,
|
| 265 |
+
"loss": 0.0474639892578125,
|
| 266 |
+
"step": 170
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2952340784479123,
|
| 270 |
+
"grad_norm": 0.06530184393433881,
|
| 271 |
+
"learning_rate": 5.94857514958738e-06,
|
| 272 |
+
"loss": 0.043914794921875,
|
| 273 |
+
"step": 175
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.30366933783213834,
|
| 277 |
+
"grad_norm": 0.06214586771199744,
|
| 278 |
+
"learning_rate": 5.945603646427567e-06,
|
| 279 |
+
"loss": 0.043475341796875,
|
| 280 |
+
"step": 180
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.3121045972163644,
|
| 284 |
+
"grad_norm": 0.06764874450241058,
|
| 285 |
+
"learning_rate": 5.9425494702764575e-06,
|
| 286 |
+
"loss": 0.04755859375,
|
| 287 |
+
"step": 185
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.3205398566005905,
|
| 291 |
+
"grad_norm": 0.06523200399348678,
|
| 292 |
+
"learning_rate": 5.939412706854299e-06,
|
| 293 |
+
"loss": 0.044635009765625,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.32897511598481655,
|
| 298 |
+
"grad_norm": 0.060102318432770876,
|
| 299 |
+
"learning_rate": 5.9361934441992835e-06,
|
| 300 |
+
"loss": 0.042364501953125,
|
| 301 |
+
"step": 195
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.3374103753690426,
|
| 305 |
+
"grad_norm": 0.06678207500644712,
|
| 306 |
+
"learning_rate": 5.9328917726650706e-06,
|
| 307 |
+
"loss": 0.04183349609375,
|
| 308 |
+
"step": 200
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.3374103753690426,
|
| 312 |
+
"eval_loss": 0.010425936430692673,
|
| 313 |
+
"eval_margin": -0.016463442112229044,
|
| 314 |
+
"eval_mean_neg": 0.5932909250259399,
|
| 315 |
+
"eval_mean_pos": 0.8196097016334534,
|
| 316 |
+
"eval_runtime": 365.6666,
|
| 317 |
+
"eval_samples_per_second": 21.848,
|
| 318 |
+
"eval_steps_per_second": 0.342,
|
| 319 |
+
"step": 200
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.3458456347532687,
|
| 323 |
+
"grad_norm": 0.05626492604909855,
|
| 324 |
+
"learning_rate": 5.929507784918257e-06,
|
| 325 |
+
"loss": 0.040447998046875,
|
| 326 |
+
"step": 205
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.35428089413749475,
|
| 330 |
+
"grad_norm": 0.054176681030320105,
|
| 331 |
+
"learning_rate": 5.926041575935772e-06,
|
| 332 |
+
"loss": 0.037396240234375,
|
| 333 |
+
"step": 210
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.3627161535217208,
|
| 337 |
+
"grad_norm": 0.06953999336709471,
|
| 338 |
+
"learning_rate": 5.922493243002212e-06,
|
| 339 |
+
"loss": 0.042828369140625,
|
| 340 |
+
"step": 215
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.37115141290594683,
|
| 344 |
+
"grad_norm": 0.05589129508252642,
|
| 345 |
+
"learning_rate": 5.918862885707113e-06,
|
| 346 |
+
"loss": 0.034979248046875,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.3795866722901729,
|
| 351 |
+
"grad_norm": 0.07078214617147234,
|
| 352 |
+
"learning_rate": 5.915150605942153e-06,
|
| 353 |
+
"loss": 0.035723876953125,
|
| 354 |
+
"step": 225
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.38802193167439897,
|
| 358 |
+
"grad_norm": 0.06815732953530805,
|
| 359 |
+
"learning_rate": 5.911356507898291e-06,
|
| 360 |
+
"loss": 0.041973876953125,
|
| 361 |
+
"step": 230
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.39645719105862504,
|
| 365 |
+
"grad_norm": 0.062472935047014386,
|
| 366 |
+
"learning_rate": 5.907480698062848e-06,
|
| 367 |
+
"loss": 0.0356689453125,
|
| 368 |
+
"step": 235
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.4048924504428511,
|
| 372 |
+
"grad_norm": 0.06665756879409568,
|
| 373 |
+
"learning_rate": 5.90352328521651e-06,
|
| 374 |
+
"loss": 0.036456298828125,
|
| 375 |
+
"step": 240
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.41332770982707717,
|
| 379 |
+
"grad_norm": 0.06681598226193439,
|
| 380 |
+
"learning_rate": 5.899484380430284e-06,
|
| 381 |
+
"loss": 0.0343994140625,
|
| 382 |
+
"step": 245
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.42176296921130324,
|
| 386 |
+
"grad_norm": 0.060740413400477374,
|
| 387 |
+
"learning_rate": 5.895364097062374e-06,
|
| 388 |
+
"loss": 0.0318511962890625,
|
| 389 |
+
"step": 250
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.4301982285955293,
|
| 393 |
+
"grad_norm": 0.06545743307605277,
|
| 394 |
+
"learning_rate": 5.8911625507550015e-06,
|
| 395 |
+
"loss": 0.034765625,
|
| 396 |
+
"step": 255
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.4386334879797554,
|
| 400 |
+
"grad_norm": 0.0673664786591912,
|
| 401 |
+
"learning_rate": 5.88687985943116e-06,
|
| 402 |
+
"loss": 0.03580322265625,
|
| 403 |
+
"step": 260
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.44706874736398144,
|
| 407 |
+
"grad_norm": 0.06937993663032453,
|
| 408 |
+
"learning_rate": 5.882516143291308e-06,
|
| 409 |
+
"loss": 0.036236572265625,
|
| 410 |
+
"step": 265
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.4555040067482075,
|
| 414 |
+
"grad_norm": 0.0639250177544625,
|
| 415 |
+
"learning_rate": 5.878071524809988e-06,
|
| 416 |
+
"loss": 0.0317962646484375,
|
| 417 |
+
"step": 270
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.4639392661324336,
|
| 421 |
+
"grad_norm": 0.06037822600018219,
|
| 422 |
+
"learning_rate": 5.873546128732399e-06,
|
| 423 |
+
"loss": 0.0323699951171875,
|
| 424 |
+
"step": 275
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.47237452551665965,
|
| 428 |
+
"grad_norm": 0.060357976056049485,
|
| 429 |
+
"learning_rate": 5.868940082070885e-06,
|
| 430 |
+
"loss": 0.033660888671875,
|
| 431 |
+
"step": 280
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.4808097849008857,
|
| 435 |
+
"grad_norm": 0.061037172126093234,
|
| 436 |
+
"learning_rate": 5.8642535141013785e-06,
|
| 437 |
+
"loss": 0.0297515869140625,
|
| 438 |
+
"step": 285
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.4892450442851118,
|
| 442 |
+
"grad_norm": 0.0524126813526148,
|
| 443 |
+
"learning_rate": 5.859486556359768e-06,
|
| 444 |
+
"loss": 0.028472900390625,
|
| 445 |
+
"step": 290
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.49768030366933785,
|
| 449 |
+
"grad_norm": 0.062026009465912704,
|
| 450 |
+
"learning_rate": 5.854639342638208e-06,
|
| 451 |
+
"loss": 0.030718994140625,
|
| 452 |
+
"step": 295
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.5061155630535639,
|
| 456 |
+
"grad_norm": 0.05866098788599579,
|
| 457 |
+
"learning_rate": 5.849712008981361e-06,
|
| 458 |
+
"loss": 0.032916259765625,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.5061155630535639,
|
| 463 |
+
"eval_loss": 0.007868120446801186,
|
| 464 |
+
"eval_margin": -0.013977996595654517,
|
| 465 |
+
"eval_mean_neg": 0.5548827648162842,
|
| 466 |
+
"eval_mean_pos": 0.793705403804779,
|
| 467 |
+
"eval_runtime": 364.6437,
|
| 468 |
+
"eval_samples_per_second": 21.909,
|
| 469 |
+
"eval_steps_per_second": 0.343,
|
| 470 |
+
"step": 300
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.5145508224377899,
|
| 474 |
+
"grad_norm": 0.05968132039231295,
|
| 475 |
+
"learning_rate": 5.844704693682583e-06,
|
| 476 |
+
"loss": 0.0292724609375,
|
| 477 |
+
"step": 305
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.522986081822016,
|
| 481 |
+
"grad_norm": 0.06038138238675174,
|
| 482 |
+
"learning_rate": 5.8396175372800405e-06,
|
| 483 |
+
"loss": 0.030743408203125,
|
| 484 |
+
"step": 310
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.5314213412062421,
|
| 488 |
+
"grad_norm": 0.06052295196543659,
|
| 489 |
+
"learning_rate": 5.834450682552765e-06,
|
| 490 |
+
"loss": 0.030194091796875,
|
| 491 |
+
"step": 315
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.5398566005904681,
|
| 495 |
+
"grad_norm": 0.05539528727202974,
|
| 496 |
+
"learning_rate": 5.829204274516648e-06,
|
| 497 |
+
"loss": 0.0312774658203125,
|
| 498 |
+
"step": 320
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.5482918599746942,
|
| 502 |
+
"grad_norm": 0.052508369724972796,
|
| 503 |
+
"learning_rate": 5.823878460420366e-06,
|
| 504 |
+
"loss": 0.0295318603515625,
|
| 505 |
+
"step": 325
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.5567271193589203,
|
| 509 |
+
"grad_norm": 0.05151880865825463,
|
| 510 |
+
"learning_rate": 5.8184733897412565e-06,
|
| 511 |
+
"loss": 0.028912353515625,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.5651623787431463,
|
| 516 |
+
"grad_norm": 0.0624220665428448,
|
| 517 |
+
"learning_rate": 5.812989214181113e-06,
|
| 518 |
+
"loss": 0.027313232421875,
|
| 519 |
+
"step": 335
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.5735976381273724,
|
| 523 |
+
"grad_norm": 0.06481057308539884,
|
| 524 |
+
"learning_rate": 5.807426087661934e-06,
|
| 525 |
+
"loss": 0.02608642578125,
|
| 526 |
+
"step": 340
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.5820328975115985,
|
| 530 |
+
"grad_norm": 0.06109467057046473,
|
| 531 |
+
"learning_rate": 5.8017841663216e-06,
|
| 532 |
+
"loss": 0.0282989501953125,
|
| 533 |
+
"step": 345
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.5904681568958245,
|
| 537 |
+
"grad_norm": 0.062107444796084835,
|
| 538 |
+
"learning_rate": 5.796063608509493e-06,
|
| 539 |
+
"loss": 0.0277069091796875,
|
| 540 |
+
"step": 350
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.5989034162800506,
|
| 544 |
+
"grad_norm": 0.0552072139581444,
|
| 545 |
+
"learning_rate": 5.7902645747820485e-06,
|
| 546 |
+
"loss": 0.028399658203125,
|
| 547 |
+
"step": 355
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.6073386756642767,
|
| 551 |
+
"grad_norm": 0.06047980839414296,
|
| 552 |
+
"learning_rate": 5.784387227898254e-06,
|
| 553 |
+
"loss": 0.0281524658203125,
|
| 554 |
+
"step": 360
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.6157739350485028,
|
| 558 |
+
"grad_norm": 0.05336288606895412,
|
| 559 |
+
"learning_rate": 5.778431732815078e-06,
|
| 560 |
+
"loss": 0.02484130859375,
|
| 561 |
+
"step": 365
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.6242091944327288,
|
| 565 |
+
"grad_norm": 0.060745200996401724,
|
| 566 |
+
"learning_rate": 5.77239825668284e-06,
|
| 567 |
+
"loss": 0.02640380859375,
|
| 568 |
+
"step": 370
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.6326444538169549,
|
| 572 |
+
"grad_norm": 0.048268694566304324,
|
| 573 |
+
"learning_rate": 5.766286968840522e-06,
|
| 574 |
+
"loss": 0.0278717041015625,
|
| 575 |
+
"step": 375
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.641079713201181,
|
| 579 |
+
"grad_norm": 0.05424806603710711,
|
| 580 |
+
"learning_rate": 5.760098040811012e-06,
|
| 581 |
+
"loss": 0.0271453857421875,
|
| 582 |
+
"step": 380
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.649514972585407,
|
| 586 |
+
"grad_norm": 0.054535443289609395,
|
| 587 |
+
"learning_rate": 5.7538316462962935e-06,
|
| 588 |
+
"loss": 0.026611328125,
|
| 589 |
+
"step": 385
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.6579502319696331,
|
| 593 |
+
"grad_norm": 0.06967389025087475,
|
| 594 |
+
"learning_rate": 5.7474879611725655e-06,
|
| 595 |
+
"loss": 0.02589111328125,
|
| 596 |
+
"step": 390
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.6663854913538592,
|
| 600 |
+
"grad_norm": 0.06024092137696802,
|
| 601 |
+
"learning_rate": 5.741067163485314e-06,
|
| 602 |
+
"loss": 0.0193756103515625,
|
| 603 |
+
"step": 395
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.6748207507380852,
|
| 607 |
+
"grad_norm": 0.05981804001044263,
|
| 608 |
+
"learning_rate": 5.7345694334443066e-06,
|
| 609 |
+
"loss": 0.0205718994140625,
|
| 610 |
+
"step": 400
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.6748207507380852,
|
| 614 |
+
"eval_loss": 0.006414474919438362,
|
| 615 |
+
"eval_margin": -0.013447051244457402,
|
| 616 |
+
"eval_mean_neg": 0.5324161052703857,
|
| 617 |
+
"eval_mean_pos": 0.7909372448921204,
|
| 618 |
+
"eval_runtime": 365.0145,
|
| 619 |
+
"eval_samples_per_second": 21.887,
|
| 620 |
+
"eval_steps_per_second": 0.342,
|
| 621 |
+
"step": 400
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.6832560101223113,
|
| 625 |
+
"grad_norm": 0.06692561927901217,
|
| 626 |
+
"learning_rate": 5.727994953418538e-06,
|
| 627 |
+
"loss": 0.022021484375,
|
| 628 |
+
"step": 405
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.6916912695065374,
|
| 632 |
+
"grad_norm": 0.06609269963808409,
|
| 633 |
+
"learning_rate": 5.721343907931114e-06,
|
| 634 |
+
"loss": 0.02950592041015625,
|
| 635 |
+
"step": 410
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.7001265288907634,
|
| 639 |
+
"grad_norm": 0.052617111357424175,
|
| 640 |
+
"learning_rate": 5.71461648365407e-06,
|
| 641 |
+
"loss": 0.025189208984375,
|
| 642 |
+
"step": 415
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.7085617882749895,
|
| 646 |
+
"grad_norm": 0.04860971480260525,
|
| 647 |
+
"learning_rate": 5.707812869403128e-06,
|
| 648 |
+
"loss": 0.022052001953125,
|
| 649 |
+
"step": 420
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.7169970476592156,
|
| 653 |
+
"grad_norm": 0.06030454097987917,
|
| 654 |
+
"learning_rate": 5.7009332561324085e-06,
|
| 655 |
+
"loss": 0.0219390869140625,
|
| 656 |
+
"step": 425
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.7254323070434416,
|
| 660 |
+
"grad_norm": 0.06837586048390999,
|
| 661 |
+
"learning_rate": 5.693977836929057e-06,
|
| 662 |
+
"loss": 0.0270172119140625,
|
| 663 |
+
"step": 430
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.7338675664276677,
|
| 667 |
+
"grad_norm": 0.05197492190608033,
|
| 668 |
+
"learning_rate": 5.686946807007834e-06,
|
| 669 |
+
"loss": 0.02206878662109375,
|
| 670 |
+
"step": 435
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.7423028258118937,
|
| 674 |
+
"grad_norm": 0.06128713786873146,
|
| 675 |
+
"learning_rate": 5.679840363705637e-06,
|
| 676 |
+
"loss": 0.0244720458984375,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.7507380851961197,
|
| 681 |
+
"grad_norm": 0.0629198604819534,
|
| 682 |
+
"learning_rate": 5.672658706475953e-06,
|
| 683 |
+
"loss": 0.0194488525390625,
|
| 684 |
+
"step": 445
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.7591733445803458,
|
| 688 |
+
"grad_norm": 0.05502172045134509,
|
| 689 |
+
"learning_rate": 5.665402036883267e-06,
|
| 690 |
+
"loss": 0.0225250244140625,
|
| 691 |
+
"step": 450
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.7676086039645719,
|
| 695 |
+
"grad_norm": 0.06119000768724386,
|
| 696 |
+
"learning_rate": 5.658070558597408e-06,
|
| 697 |
+
"loss": 0.01928558349609375,
|
| 698 |
+
"step": 455
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.7760438633487979,
|
| 702 |
+
"grad_norm": 0.058834092769235756,
|
| 703 |
+
"learning_rate": 5.650664477387824e-06,
|
| 704 |
+
"loss": 0.02149658203125,
|
| 705 |
+
"step": 460
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.784479122733024,
|
| 709 |
+
"grad_norm": 0.06942758384696321,
|
| 710 |
+
"learning_rate": 5.643184001117811e-06,
|
| 711 |
+
"loss": 0.0266326904296875,
|
| 712 |
+
"step": 465
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.7929143821172501,
|
| 716 |
+
"grad_norm": 0.05395397336586372,
|
| 717 |
+
"learning_rate": 5.6356293397386836e-06,
|
| 718 |
+
"loss": 0.0206085205078125,
|
| 719 |
+
"step": 470
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.8013496415014761,
|
| 723 |
+
"grad_norm": 0.057301086470950384,
|
| 724 |
+
"learning_rate": 5.628000705283873e-06,
|
| 725 |
+
"loss": 0.021770477294921875,
|
| 726 |
+
"step": 475
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.8097849008857022,
|
| 730 |
+
"grad_norm": 0.058618795566843934,
|
| 731 |
+
"learning_rate": 5.620298311862985e-06,
|
| 732 |
+
"loss": 0.0174072265625,
|
| 733 |
+
"step": 480
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.8182201602699283,
|
| 737 |
+
"grad_norm": 0.053997897902853975,
|
| 738 |
+
"learning_rate": 5.612522375655783e-06,
|
| 739 |
+
"loss": 0.0246124267578125,
|
| 740 |
+
"step": 485
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.8266554196541543,
|
| 744 |
+
"grad_norm": 0.058293384553658546,
|
| 745 |
+
"learning_rate": 5.604673114906126e-06,
|
| 746 |
+
"loss": 0.0239288330078125,
|
| 747 |
+
"step": 490
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.8350906790383804,
|
| 751 |
+
"grad_norm": 0.062099166751088966,
|
| 752 |
+
"learning_rate": 5.596750749915842e-06,
|
| 753 |
+
"loss": 0.023724365234375,
|
| 754 |
+
"step": 495
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.8435259384226065,
|
| 758 |
+
"grad_norm": 0.0518337334475497,
|
| 759 |
+
"learning_rate": 5.588755503038543e-06,
|
| 760 |
+
"loss": 0.01995849609375,
|
| 761 |
+
"step": 500
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.8435259384226065,
|
| 765 |
+
"eval_loss": 0.00542406877502799,
|
| 766 |
+
"eval_margin": -0.010786364688688228,
|
| 767 |
+
"eval_mean_neg": 0.5346763134002686,
|
| 768 |
+
"eval_mean_pos": 0.7906754016876221,
|
| 769 |
+
"eval_runtime": 367.5462,
|
| 770 |
+
"eval_samples_per_second": 21.736,
|
| 771 |
+
"eval_steps_per_second": 0.34,
|
| 772 |
+
"step": 500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.8519611978068325,
|
| 776 |
+
"grad_norm": 0.0579368996460804,
|
| 777 |
+
"learning_rate": 5.580687598673387e-06,
|
| 778 |
+
"loss": 0.02121734619140625,
|
| 779 |
+
"step": 505
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.8603964571910586,
|
| 783 |
+
"grad_norm": 0.05214070956939639,
|
| 784 |
+
"learning_rate": 5.572547263258776e-06,
|
| 785 |
+
"loss": 0.0197113037109375,
|
| 786 |
+
"step": 510
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.8688317165752847,
|
| 790 |
+
"grad_norm": 0.0687906199565583,
|
| 791 |
+
"learning_rate": 5.564334725266006e-06,
|
| 792 |
+
"loss": 0.0217254638671875,
|
| 793 |
+
"step": 515
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.8772669759595108,
|
| 797 |
+
"grad_norm": 0.051621267659708626,
|
| 798 |
+
"learning_rate": 5.55605021519285e-06,
|
| 799 |
+
"loss": 0.019158935546875,
|
| 800 |
+
"step": 520
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.8857022353437368,
|
| 804 |
+
"grad_norm": 0.05599957001213385,
|
| 805 |
+
"learning_rate": 5.547693965557092e-06,
|
| 806 |
+
"loss": 0.0195770263671875,
|
| 807 |
+
"step": 525
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.8941374947279629,
|
| 811 |
+
"grad_norm": 0.06034671456944424,
|
| 812 |
+
"learning_rate": 5.539266210889997e-06,
|
| 813 |
+
"loss": 0.0231231689453125,
|
| 814 |
+
"step": 530
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.902572754112189,
|
| 818 |
+
"grad_norm": 0.04518349407201743,
|
| 819 |
+
"learning_rate": 5.5307671877297326e-06,
|
| 820 |
+
"loss": 0.0208709716796875,
|
| 821 |
+
"step": 535
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.911008013496415,
|
| 825 |
+
"grad_norm": 0.0503478793140038,
|
| 826 |
+
"learning_rate": 5.522197134614728e-06,
|
| 827 |
+
"loss": 0.0209930419921875,
|
| 828 |
+
"step": 540
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.9194432728806411,
|
| 832 |
+
"grad_norm": 0.046047217532892024,
|
| 833 |
+
"learning_rate": 5.513556292076981e-06,
|
| 834 |
+
"loss": 0.0175750732421875,
|
| 835 |
+
"step": 545
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.9278785322648672,
|
| 839 |
+
"grad_norm": 0.05575253759567789,
|
| 840 |
+
"learning_rate": 5.504844902635303e-06,
|
| 841 |
+
"loss": 0.0171112060546875,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.9363137916490932,
|
| 846 |
+
"grad_norm": 0.04687503220455111,
|
| 847 |
+
"learning_rate": 5.496063210788519e-06,
|
| 848 |
+
"loss": 0.0167633056640625,
|
| 849 |
+
"step": 555
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.9447490510333193,
|
| 853 |
+
"grad_norm": 0.04891593875536363,
|
| 854 |
+
"learning_rate": 5.487211463008597e-06,
|
| 855 |
+
"loss": 0.019036865234375,
|
| 856 |
+
"step": 560
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.9531843104175454,
|
| 860 |
+
"grad_norm": 0.04841249311058062,
|
| 861 |
+
"learning_rate": 5.478289907733738e-06,
|
| 862 |
+
"loss": 0.01807098388671875,
|
| 863 |
+
"step": 565
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.9616195698017714,
|
| 867 |
+
"grad_norm": 0.060373651634708765,
|
| 868 |
+
"learning_rate": 5.469298795361397e-06,
|
| 869 |
+
"loss": 0.015673828125,
|
| 870 |
+
"step": 570
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.9700548291859975,
|
| 874 |
+
"grad_norm": 0.051868174671481436,
|
| 875 |
+
"learning_rate": 5.460238378241262e-06,
|
| 876 |
+
"loss": 0.01802978515625,
|
| 877 |
+
"step": 575
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.9784900885702236,
|
| 881 |
+
"grad_norm": 0.051146316151485995,
|
| 882 |
+
"learning_rate": 5.451108910668163e-06,
|
| 883 |
+
"loss": 0.01664581298828125,
|
| 884 |
+
"step": 580
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.9869253479544496,
|
| 888 |
+
"grad_norm": 0.04017649470362814,
|
| 889 |
+
"learning_rate": 5.441910648874945e-06,
|
| 890 |
+
"loss": 0.016483306884765625,
|
| 891 |
+
"step": 585
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.9953606073386757,
|
| 895 |
+
"grad_norm": 0.04457228909606784,
|
| 896 |
+
"learning_rate": 5.4326438510252655e-06,
|
| 897 |
+
"loss": 0.0192718505859375,
|
| 898 |
+
"step": 590
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.0033741037536905,
|
| 902 |
+
"grad_norm": 0.053005736672298354,
|
| 903 |
+
"learning_rate": 5.423308777206357e-06,
|
| 904 |
+
"loss": 0.015604400634765625,
|
| 905 |
+
"step": 595
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.0118093631379166,
|
| 909 |
+
"grad_norm": 0.05754347568157857,
|
| 910 |
+
"learning_rate": 5.413905689421722e-06,
|
| 911 |
+
"loss": 0.0159515380859375,
|
| 912 |
+
"step": 600
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.0118093631379166,
|
| 916 |
+
"eval_loss": 0.004593910649418831,
|
| 917 |
+
"eval_margin": -0.011103880922159842,
|
| 918 |
+
"eval_mean_neg": 0.5064941644668579,
|
| 919 |
+
"eval_mean_pos": 0.7831713557243347,
|
| 920 |
+
"eval_runtime": 364.5429,
|
| 921 |
+
"eval_samples_per_second": 21.915,
|
| 922 |
+
"eval_steps_per_second": 0.343,
|
| 923 |
+
"step": 600
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.0202446225221427,
|
| 927 |
+
"grad_norm": 0.04962686662442784,
|
| 928 |
+
"learning_rate": 5.404434851583785e-06,
|
| 929 |
+
"loss": 0.01360015869140625,
|
| 930 |
+
"step": 605
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 1.0286798819063687,
|
| 934 |
+
"grad_norm": 0.046926535788142015,
|
| 935 |
+
"learning_rate": 5.394896529506479e-06,
|
| 936 |
+
"loss": 0.01566925048828125,
|
| 937 |
+
"step": 610
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 1.0371151412905948,
|
| 941 |
+
"grad_norm": 0.043042108440633,
|
| 942 |
+
"learning_rate": 5.38529099089779e-06,
|
| 943 |
+
"loss": 0.0128326416015625,
|
| 944 |
+
"step": 615
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.0455504006748209,
|
| 948 |
+
"grad_norm": 0.049749099013614635,
|
| 949 |
+
"learning_rate": 5.375618505352241e-06,
|
| 950 |
+
"loss": 0.0136383056640625,
|
| 951 |
+
"step": 620
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 1.053985660059047,
|
| 955 |
+
"grad_norm": 0.04945151693616336,
|
| 956 |
+
"learning_rate": 5.365879344343326e-06,
|
| 957 |
+
"loss": 0.01544036865234375,
|
| 958 |
+
"step": 625
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.062420919443273,
|
| 962 |
+
"grad_norm": 0.04626935309793636,
|
| 963 |
+
"learning_rate": 5.35607378121589e-06,
|
| 964 |
+
"loss": 0.0143829345703125,
|
| 965 |
+
"step": 630
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 1.070856178827499,
|
| 969 |
+
"grad_norm": 0.04580735975264899,
|
| 970 |
+
"learning_rate": 5.346202091178459e-06,
|
| 971 |
+
"loss": 0.014122772216796874,
|
| 972 |
+
"step": 635
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 1.079291438211725,
|
| 976 |
+
"grad_norm": 0.046216725385350446,
|
| 977 |
+
"learning_rate": 5.336264551295512e-06,
|
| 978 |
+
"loss": 0.014672088623046874,
|
| 979 |
+
"step": 640
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 1.087726697595951,
|
| 983 |
+
"grad_norm": 0.04564303944680029,
|
| 984 |
+
"learning_rate": 5.326261440479709e-06,
|
| 985 |
+
"loss": 0.0136993408203125,
|
| 986 |
+
"step": 645
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.096161956980177,
|
| 990 |
+
"grad_norm": 0.05114495970312972,
|
| 991 |
+
"learning_rate": 5.316193039484063e-06,
|
| 992 |
+
"loss": 0.0147705078125,
|
| 993 |
+
"step": 650
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 1.1045972163644031,
|
| 997 |
+
"grad_norm": 0.044105955284847585,
|
| 998 |
+
"learning_rate": 5.306059630894056e-06,
|
| 999 |
+
"loss": 0.015480804443359374,
|
| 1000 |
+
"step": 655
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.1130324757486292,
|
| 1004 |
+
"grad_norm": 0.045182200484827885,
|
| 1005 |
+
"learning_rate": 5.295861499119711e-06,
|
| 1006 |
+
"loss": 0.013404083251953126,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.1214677351328552,
|
| 1011 |
+
"grad_norm": 0.04324759296793784,
|
| 1012 |
+
"learning_rate": 5.2855989303876065e-06,
|
| 1013 |
+
"loss": 0.01672821044921875,
|
| 1014 |
+
"step": 665
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 1.1299029945170813,
|
| 1018 |
+
"grad_norm": 0.03328038907845692,
|
| 1019 |
+
"learning_rate": 5.275272212732849e-06,
|
| 1020 |
+
"loss": 0.01335906982421875,
|
| 1021 |
+
"step": 670
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.1383382539013074,
|
| 1025 |
+
"grad_norm": 0.044225327184826406,
|
| 1026 |
+
"learning_rate": 5.264881635990984e-06,
|
| 1027 |
+
"loss": 0.012935638427734375,
|
| 1028 |
+
"step": 675
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.1467735132855335,
|
| 1032 |
+
"grad_norm": 0.04645591264342837,
|
| 1033 |
+
"learning_rate": 5.2544274917898615e-06,
|
| 1034 |
+
"loss": 0.01385498046875,
|
| 1035 |
+
"step": 680
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 1.1552087726697595,
|
| 1039 |
+
"grad_norm": 0.05453216622664439,
|
| 1040 |
+
"learning_rate": 5.243910073541454e-06,
|
| 1041 |
+
"loss": 0.016290283203125,
|
| 1042 |
+
"step": 685
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 1.1636440320539856,
|
| 1046 |
+
"grad_norm": 0.057731965028177075,
|
| 1047 |
+
"learning_rate": 5.233329676433617e-06,
|
| 1048 |
+
"loss": 0.0145355224609375,
|
| 1049 |
+
"step": 690
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 1.1720792914382117,
|
| 1053 |
+
"grad_norm": 0.05145183297720149,
|
| 1054 |
+
"learning_rate": 5.222686597421808e-06,
|
| 1055 |
+
"loss": 0.01390838623046875,
|
| 1056 |
+
"step": 695
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 1.1805145508224377,
|
| 1060 |
+
"grad_norm": 0.04021056012812571,
|
| 1061 |
+
"learning_rate": 5.211981135220751e-06,
|
| 1062 |
+
"loss": 0.01344757080078125,
|
| 1063 |
+
"step": 700
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 1.1805145508224377,
|
| 1067 |
+
"eval_loss": 0.004191060084849596,
|
| 1068 |
+
"eval_margin": -0.010819014589933137,
|
| 1069 |
+
"eval_mean_neg": 0.4851545989513397,
|
| 1070 |
+
"eval_mean_pos": 0.7733471989631653,
|
| 1071 |
+
"eval_runtime": 359.0481,
|
| 1072 |
+
"eval_samples_per_second": 22.251,
|
| 1073 |
+
"eval_steps_per_second": 0.348,
|
| 1074 |
+
"step": 700
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.1889498102066638,
|
| 1078 |
+
"grad_norm": 0.0416204676277527,
|
| 1079 |
+
"learning_rate": 5.201213590296052e-06,
|
| 1080 |
+
"loss": 0.014748382568359374,
|
| 1081 |
+
"step": 705
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.1973850695908899,
|
| 1085 |
+
"grad_norm": 0.05633713089091016,
|
| 1086 |
+
"learning_rate": 5.190384264855764e-06,
|
| 1087 |
+
"loss": 0.014013671875,
|
| 1088 |
+
"step": 710
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.205820328975116,
|
| 1092 |
+
"grad_norm": 0.05143948467095745,
|
| 1093 |
+
"learning_rate": 5.1794934628419104e-06,
|
| 1094 |
+
"loss": 0.015460205078125,
|
| 1095 |
+
"step": 715
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.214255588359342,
|
| 1099 |
+
"grad_norm": 0.05227911954680101,
|
| 1100 |
+
"learning_rate": 5.168541489921949e-06,
|
| 1101 |
+
"loss": 0.01507415771484375,
|
| 1102 |
+
"step": 720
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.222690847743568,
|
| 1106 |
+
"grad_norm": 0.058608960783147375,
|
| 1107 |
+
"learning_rate": 5.1575286534801955e-06,
|
| 1108 |
+
"loss": 0.01417236328125,
|
| 1109 |
+
"step": 725
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2311261071277941,
|
| 1113 |
+
"grad_norm": 0.04818858161693878,
|
| 1114 |
+
"learning_rate": 5.146455262609197e-06,
|
| 1115 |
+
"loss": 0.013425445556640625,
|
| 1116 |
+
"step": 730
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2395613665120202,
|
| 1120 |
+
"grad_norm": 0.05406749848988645,
|
| 1121 |
+
"learning_rate": 5.1353216281010535e-06,
|
| 1122 |
+
"loss": 0.013022613525390626,
|
| 1123 |
+
"step": 735
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.2479966258962463,
|
| 1127 |
+
"grad_norm": 0.044408669007062154,
|
| 1128 |
+
"learning_rate": 5.1241280624387e-06,
|
| 1129 |
+
"loss": 0.01393585205078125,
|
| 1130 |
+
"step": 740
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2564318852804723,
|
| 1134 |
+
"grad_norm": 0.04519048638967848,
|
| 1135 |
+
"learning_rate": 5.1128748797871314e-06,
|
| 1136 |
+
"loss": 0.013826751708984375,
|
| 1137 |
+
"step": 745
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.2648671446646984,
|
| 1141 |
+
"grad_norm": 0.0491460974626283,
|
| 1142 |
+
"learning_rate": 5.101562395984587e-06,
|
| 1143 |
+
"loss": 0.01336212158203125,
|
| 1144 |
+
"step": 750
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.2733024040489245,
|
| 1148 |
+
"grad_norm": 0.04356609182045035,
|
| 1149 |
+
"learning_rate": 5.090190928533689e-06,
|
| 1150 |
+
"loss": 0.01492156982421875,
|
| 1151 |
+
"step": 755
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.2817376634331505,
|
| 1155 |
+
"grad_norm": 0.03556136795064142,
|
| 1156 |
+
"learning_rate": 5.078760796592524e-06,
|
| 1157 |
+
"loss": 0.0125732421875,
|
| 1158 |
+
"step": 760
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.2901729228173766,
|
| 1162 |
+
"grad_norm": 0.04189977738590891,
|
| 1163 |
+
"learning_rate": 5.067272320965692e-06,
|
| 1164 |
+
"loss": 0.0149322509765625,
|
| 1165 |
+
"step": 765
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.2986081822016027,
|
| 1169 |
+
"grad_norm": 0.05051201336701144,
|
| 1170 |
+
"learning_rate": 5.055725824095301e-06,
|
| 1171 |
+
"loss": 0.01419525146484375,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.3070434415858287,
|
| 1176 |
+
"grad_norm": 0.0416942039130722,
|
| 1177 |
+
"learning_rate": 5.0441216300519126e-06,
|
| 1178 |
+
"loss": 0.01274261474609375,
|
| 1179 |
+
"step": 775
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.3154787009700548,
|
| 1183 |
+
"grad_norm": 0.04629875001130603,
|
| 1184 |
+
"learning_rate": 5.032460064525455e-06,
|
| 1185 |
+
"loss": 0.01363525390625,
|
| 1186 |
+
"step": 780
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3239139603542809,
|
| 1190 |
+
"grad_norm": 0.03704688355237128,
|
| 1191 |
+
"learning_rate": 5.020741454816074e-06,
|
| 1192 |
+
"loss": 0.01301422119140625,
|
| 1193 |
+
"step": 785
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.332349219738507,
|
| 1197 |
+
"grad_norm": 0.03742406408262459,
|
| 1198 |
+
"learning_rate": 5.00896612982495e-06,
|
| 1199 |
+
"loss": 0.01353302001953125,
|
| 1200 |
+
"step": 790
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.340784479122733,
|
| 1204 |
+
"grad_norm": 0.050480726423335516,
|
| 1205 |
+
"learning_rate": 4.99713442004507e-06,
|
| 1206 |
+
"loss": 0.01196746826171875,
|
| 1207 |
+
"step": 795
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.349219738506959,
|
| 1211 |
+
"grad_norm": 0.03808846024736694,
|
| 1212 |
+
"learning_rate": 4.985246657551943e-06,
|
| 1213 |
+
"loss": 0.0110015869140625,
|
| 1214 |
+
"step": 800
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.349219738506959,
|
| 1218 |
+
"eval_loss": 0.003908403683453798,
|
| 1219 |
+
"eval_margin": -0.010123856463319352,
|
| 1220 |
+
"eval_mean_neg": 0.49688851833343506,
|
| 1221 |
+
"eval_mean_pos": 0.7784863114356995,
|
| 1222 |
+
"eval_runtime": 364.0137,
|
| 1223 |
+
"eval_samples_per_second": 21.947,
|
| 1224 |
+
"eval_steps_per_second": 0.343,
|
| 1225 |
+
"step": 800
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 1.3576549978911852,
|
| 1229 |
+
"grad_norm": 0.04637758927467518,
|
| 1230 |
+
"learning_rate": 4.973303175994289e-06,
|
| 1231 |
+
"loss": 0.013458251953125,
|
| 1232 |
+
"step": 805
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.3660902572754112,
|
| 1236 |
+
"grad_norm": 0.05066098296531039,
|
| 1237 |
+
"learning_rate": 4.961304310584674e-06,
|
| 1238 |
+
"loss": 0.01515960693359375,
|
| 1239 |
+
"step": 810
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.3745255166596373,
|
| 1243 |
+
"grad_norm": 0.038530384714911596,
|
| 1244 |
+
"learning_rate": 4.949250398090092e-06,
|
| 1245 |
+
"loss": 0.011260223388671876,
|
| 1246 |
+
"step": 815
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.3829607760438634,
|
| 1250 |
+
"grad_norm": 0.040188601844867354,
|
| 1251 |
+
"learning_rate": 4.937141776822525e-06,
|
| 1252 |
+
"loss": 0.0158447265625,
|
| 1253 |
+
"step": 820
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 1.3913960354280894,
|
| 1257 |
+
"grad_norm": 0.03574613677300634,
|
| 1258 |
+
"learning_rate": 4.92497878662944e-06,
|
| 1259 |
+
"loss": 0.011143875122070313,
|
| 1260 |
+
"step": 825
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 1.3998312948123155,
|
| 1264 |
+
"grad_norm": 0.05019423126073816,
|
| 1265 |
+
"learning_rate": 4.912761768884255e-06,
|
| 1266 |
+
"loss": 0.01179351806640625,
|
| 1267 |
+
"step": 830
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 1.4082665541965416,
|
| 1271 |
+
"grad_norm": 0.04311116805857567,
|
| 1272 |
+
"learning_rate": 4.9004910664767545e-06,
|
| 1273 |
+
"loss": 0.01372833251953125,
|
| 1274 |
+
"step": 835
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.4167018135807676,
|
| 1278 |
+
"grad_norm": 0.04928580588462512,
|
| 1279 |
+
"learning_rate": 4.888167023803468e-06,
|
| 1280 |
+
"loss": 0.01297607421875,
|
| 1281 |
+
"step": 840
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 1.4251370729649937,
|
| 1285 |
+
"grad_norm": 0.054968450905918724,
|
| 1286 |
+
"learning_rate": 4.8757899867580046e-06,
|
| 1287 |
+
"loss": 0.014654541015625,
|
| 1288 |
+
"step": 845
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 1.4335723323492198,
|
| 1292 |
+
"grad_norm": 0.050366347428194534,
|
| 1293 |
+
"learning_rate": 4.86336030272134e-06,
|
| 1294 |
+
"loss": 0.011295318603515625,
|
| 1295 |
+
"step": 850
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.4420075917334458,
|
| 1299 |
+
"grad_norm": 0.05107215089989217,
|
| 1300 |
+
"learning_rate": 4.850878320552076e-06,
|
| 1301 |
+
"loss": 0.01334228515625,
|
| 1302 |
+
"step": 855
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 1.450442851117672,
|
| 1306 |
+
"grad_norm": 0.0391963683003482,
|
| 1307 |
+
"learning_rate": 4.838344390576638e-06,
|
| 1308 |
+
"loss": 0.01104736328125,
|
| 1309 |
+
"step": 860
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 1.458878110501898,
|
| 1313 |
+
"grad_norm": 0.03985676744245212,
|
| 1314 |
+
"learning_rate": 4.825758864579452e-06,
|
| 1315 |
+
"loss": 0.013307952880859375,
|
| 1316 |
+
"step": 865
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 1.467313369886124,
|
| 1320 |
+
"grad_norm": 0.04852757651119817,
|
| 1321 |
+
"learning_rate": 4.813122095793066e-06,
|
| 1322 |
+
"loss": 0.014328384399414062,
|
| 1323 |
+
"step": 870
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 1.47574862927035,
|
| 1327 |
+
"grad_norm": 0.0454254941425111,
|
| 1328 |
+
"learning_rate": 4.800434438888235e-06,
|
| 1329 |
+
"loss": 0.012960052490234375,
|
| 1330 |
+
"step": 875
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 1.4841838886545762,
|
| 1334 |
+
"grad_norm": 0.03868230007157653,
|
| 1335 |
+
"learning_rate": 4.787696249963974e-06,
|
| 1336 |
+
"loss": 0.01402740478515625,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 1.4926191480388022,
|
| 1341 |
+
"grad_norm": 0.05289135869423979,
|
| 1342 |
+
"learning_rate": 4.774907886537553e-06,
|
| 1343 |
+
"loss": 0.013831901550292968,
|
| 1344 |
+
"step": 885
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 1.501054407423028,
|
| 1348 |
+
"grad_norm": 0.04594308680556284,
|
| 1349 |
+
"learning_rate": 4.7620697075344736e-06,
|
| 1350 |
+
"loss": 0.012446975708007813,
|
| 1351 |
+
"step": 890
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 1.5094896668072542,
|
| 1355 |
+
"grad_norm": 0.048917845490978454,
|
| 1356 |
+
"learning_rate": 4.7491820732783866e-06,
|
| 1357 |
+
"loss": 0.011295318603515625,
|
| 1358 |
+
"step": 895
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 1.5179249261914802,
|
| 1362 |
+
"grad_norm": 0.043266255463378436,
|
| 1363 |
+
"learning_rate": 4.73624534548098e-06,
|
| 1364 |
+
"loss": 0.01407012939453125,
|
| 1365 |
+
"step": 900
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 1.5179249261914802,
|
| 1369 |
+
"eval_loss": 0.0036048581823706627,
|
| 1370 |
+
"eval_margin": -0.009617562525935711,
|
| 1371 |
+
"eval_mean_neg": 0.4904225468635559,
|
| 1372 |
+
"eval_mean_pos": 0.7793014049530029,
|
| 1373 |
+
"eval_runtime": 363.5397,
|
| 1374 |
+
"eval_samples_per_second": 21.976,
|
| 1375 |
+
"eval_steps_per_second": 0.344,
|
| 1376 |
+
"step": 900
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 1.5263601855757063,
|
| 1380 |
+
"grad_norm": 0.04363576408467007,
|
| 1381 |
+
"learning_rate": 4.723259887231835e-06,
|
| 1382 |
+
"loss": 0.0138519287109375,
|
| 1383 |
+
"step": 905
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 1.5347954449599324,
|
| 1387 |
+
"grad_norm": 0.04199459687850267,
|
| 1388 |
+
"learning_rate": 4.710226062988223e-06,
|
| 1389 |
+
"loss": 0.01312255859375,
|
| 1390 |
+
"step": 910
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 1.5432307043441584,
|
| 1394 |
+
"grad_norm": 0.047436231412077354,
|
| 1395 |
+
"learning_rate": 4.697144238564889e-06,
|
| 1396 |
+
"loss": 0.01208648681640625,
|
| 1397 |
+
"step": 915
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 1.5516659637283845,
|
| 1401 |
+
"grad_norm": 0.04092453404900873,
|
| 1402 |
+
"learning_rate": 4.684014781123775e-06,
|
| 1403 |
+
"loss": 0.012505340576171874,
|
| 1404 |
+
"step": 920
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 1.5601012231126106,
|
| 1408 |
+
"grad_norm": 0.045645370405214956,
|
| 1409 |
+
"learning_rate": 4.6708380591637166e-06,
|
| 1410 |
+
"loss": 0.0120208740234375,
|
| 1411 |
+
"step": 925
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.5685364824968366,
|
| 1415 |
+
"grad_norm": 0.04911154284719614,
|
| 1416 |
+
"learning_rate": 4.6576144425101076e-06,
|
| 1417 |
+
"loss": 0.013311767578125,
|
| 1418 |
+
"step": 930
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 1.5769717418810627,
|
| 1422 |
+
"grad_norm": 0.045881762593597546,
|
| 1423 |
+
"learning_rate": 4.64434430230451e-06,
|
| 1424 |
+
"loss": 0.012969207763671876,
|
| 1425 |
+
"step": 935
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 1.5854070012652888,
|
| 1429 |
+
"grad_norm": 0.04728445094523914,
|
| 1430 |
+
"learning_rate": 4.631028010994245e-06,
|
| 1431 |
+
"loss": 0.01099395751953125,
|
| 1432 |
+
"step": 940
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 1.5938422606495148,
|
| 1436 |
+
"grad_norm": 0.03903116673162643,
|
| 1437 |
+
"learning_rate": 4.617665942321937e-06,
|
| 1438 |
+
"loss": 0.0129608154296875,
|
| 1439 |
+
"step": 945
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.602277520033741,
|
| 1443 |
+
"grad_norm": 0.040499425484585065,
|
| 1444 |
+
"learning_rate": 4.6042584713150225e-06,
|
| 1445 |
+
"loss": 0.009827423095703124,
|
| 1446 |
+
"step": 950
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 1.610712779417967,
|
| 1450 |
+
"grad_norm": 0.047017092872005554,
|
| 1451 |
+
"learning_rate": 4.590805974275228e-06,
|
| 1452 |
+
"loss": 0.01045989990234375,
|
| 1453 |
+
"step": 955
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 1.619148038802193,
|
| 1457 |
+
"grad_norm": 0.03869016761931018,
|
| 1458 |
+
"learning_rate": 4.577308828768005e-06,
|
| 1459 |
+
"loss": 0.011346435546875,
|
| 1460 |
+
"step": 960
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 1.6275832981864191,
|
| 1464 |
+
"grad_norm": 0.05726216064413269,
|
| 1465 |
+
"learning_rate": 4.563767413611932e-06,
|
| 1466 |
+
"loss": 0.01296844482421875,
|
| 1467 |
+
"step": 965
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.6360185575706452,
|
| 1471 |
+
"grad_norm": 0.034971593802495975,
|
| 1472 |
+
"learning_rate": 4.550182108868089e-06,
|
| 1473 |
+
"loss": 0.01379852294921875,
|
| 1474 |
+
"step": 970
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 1.6444538169548713,
|
| 1478 |
+
"grad_norm": 0.04877425067250454,
|
| 1479 |
+
"learning_rate": 4.536553295829384e-06,
|
| 1480 |
+
"loss": 0.012924957275390624,
|
| 1481 |
+
"step": 975
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 1.6528890763390973,
|
| 1485 |
+
"grad_norm": 0.03927648322180213,
|
| 1486 |
+
"learning_rate": 4.522881357009853e-06,
|
| 1487 |
+
"loss": 0.01293792724609375,
|
| 1488 |
+
"step": 980
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 1.6613243357233234,
|
| 1492 |
+
"grad_norm": 0.024976847462424127,
|
| 1493 |
+
"learning_rate": 4.5091666761339275e-06,
|
| 1494 |
+
"loss": 0.009877777099609375,
|
| 1495 |
+
"step": 985
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 1.6697595951075495,
|
| 1499 |
+
"grad_norm": 0.03945379802090875,
|
| 1500 |
+
"learning_rate": 4.495409638125657e-06,
|
| 1501 |
+
"loss": 0.01130523681640625,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 1.6781948544917755,
|
| 1506 |
+
"grad_norm": 0.03430320161614481,
|
| 1507 |
+
"learning_rate": 4.481610629097917e-06,
|
| 1508 |
+
"loss": 0.009923553466796875,
|
| 1509 |
+
"step": 995
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 1.6866301138760016,
|
| 1513 |
+
"grad_norm": 0.03895065600017937,
|
| 1514 |
+
"learning_rate": 4.46777003634156e-06,
|
| 1515 |
+
"loss": 0.01330413818359375,
|
| 1516 |
+
"step": 1000
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 1.6866301138760016,
|
| 1520 |
+
"eval_loss": 0.0034073551651090384,
|
| 1521 |
+
"eval_margin": -0.009528953300398444,
|
| 1522 |
+
"eval_mean_neg": 0.4926661550998688,
|
| 1523 |
+
"eval_mean_pos": 0.7842009663581848,
|
| 1524 |
+
"eval_runtime": 367.0219,
|
| 1525 |
+
"eval_samples_per_second": 21.767,
|
| 1526 |
+
"eval_steps_per_second": 0.341,
|
| 1527 |
+
"step": 1000
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 1.6950653732602277,
|
| 1531 |
+
"grad_norm": 0.04302786223265218,
|
| 1532 |
+
"learning_rate": 4.453888248314553e-06,
|
| 1533 |
+
"loss": 0.01107330322265625,
|
| 1534 |
+
"step": 1005
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 1.7035006326444537,
|
| 1538 |
+
"grad_norm": 0.04002206909489744,
|
| 1539 |
+
"learning_rate": 4.439965654631073e-06,
|
| 1540 |
+
"loss": 0.0105499267578125,
|
| 1541 |
+
"step": 1010
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 1.7119358920286798,
|
| 1545 |
+
"grad_norm": 0.04439497813433074,
|
| 1546 |
+
"learning_rate": 4.426002646050574e-06,
|
| 1547 |
+
"loss": 0.010544586181640624,
|
| 1548 |
+
"step": 1015
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 1.7203711514129059,
|
| 1552 |
+
"grad_norm": 0.043341839034531496,
|
| 1553 |
+
"learning_rate": 4.411999614466812e-06,
|
| 1554 |
+
"loss": 0.0125335693359375,
|
| 1555 |
+
"step": 1020
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 1.728806410797132,
|
| 1559 |
+
"grad_norm": 0.03449321841295583,
|
| 1560 |
+
"learning_rate": 4.397956952896858e-06,
|
| 1561 |
+
"loss": 0.010623550415039063,
|
| 1562 |
+
"step": 1025
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 1.737241670181358,
|
| 1566 |
+
"grad_norm": 0.041185961783139574,
|
| 1567 |
+
"learning_rate": 4.383875055470055e-06,
|
| 1568 |
+
"loss": 0.01031951904296875,
|
| 1569 |
+
"step": 1030
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.745676929565584,
|
| 1573 |
+
"grad_norm": 0.04627446953615271,
|
| 1574 |
+
"learning_rate": 4.3697543174169675e-06,
|
| 1575 |
+
"loss": 0.01590385437011719,
|
| 1576 |
+
"step": 1035
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.7541121889498101,
|
| 1580 |
+
"grad_norm": 0.04582345634360075,
|
| 1581 |
+
"learning_rate": 4.355595135058278e-06,
|
| 1582 |
+
"loss": 0.0119537353515625,
|
| 1583 |
+
"step": 1040
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.7625474483340362,
|
| 1587 |
+
"grad_norm": 0.033580437424405536,
|
| 1588 |
+
"learning_rate": 4.3413979057936715e-06,
|
| 1589 |
+
"loss": 0.01235198974609375,
|
| 1590 |
+
"step": 1045
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 1.7709827077182623,
|
| 1594 |
+
"grad_norm": 0.03545606353671419,
|
| 1595 |
+
"learning_rate": 4.32716302809068e-06,
|
| 1596 |
+
"loss": 0.012863922119140624,
|
| 1597 |
+
"step": 1050
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 1.7794179671024883,
|
| 1601 |
+
"grad_norm": 0.03491571698794484,
|
| 1602 |
+
"learning_rate": 4.312890901473496e-06,
|
| 1603 |
+
"loss": 0.01035614013671875,
|
| 1604 |
+
"step": 1055
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 1.7878532264867144,
|
| 1608 |
+
"grad_norm": 0.04391496148899165,
|
| 1609 |
+
"learning_rate": 4.29858192651176e-06,
|
| 1610 |
+
"loss": 0.011370468139648437,
|
| 1611 |
+
"step": 1060
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 1.7962884858709405,
|
| 1615 |
+
"grad_norm": 0.049338016603549396,
|
| 1616 |
+
"learning_rate": 4.284236504809324e-06,
|
| 1617 |
+
"loss": 0.011846160888671875,
|
| 1618 |
+
"step": 1065
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 1.8047237452551665,
|
| 1622 |
+
"grad_norm": 0.035387852478552806,
|
| 1623 |
+
"learning_rate": 4.269855038992971e-06,
|
| 1624 |
+
"loss": 0.011142349243164063,
|
| 1625 |
+
"step": 1070
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 1.8131590046393926,
|
| 1629 |
+
"grad_norm": 0.043891210942711104,
|
| 1630 |
+
"learning_rate": 4.2554379327011196e-06,
|
| 1631 |
+
"loss": 0.011545944213867187,
|
| 1632 |
+
"step": 1075
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 1.8215942640236187,
|
| 1636 |
+
"grad_norm": 0.040327331551499056,
|
| 1637 |
+
"learning_rate": 4.240985590572496e-06,
|
| 1638 |
+
"loss": 0.00897674560546875,
|
| 1639 |
+
"step": 1080
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 1.8300295234078447,
|
| 1643 |
+
"grad_norm": 0.03274271686886844,
|
| 1644 |
+
"learning_rate": 4.226498418234771e-06,
|
| 1645 |
+
"loss": 0.01215667724609375,
|
| 1646 |
+
"step": 1085
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 1.8384647827920708,
|
| 1650 |
+
"grad_norm": 0.04375742422856697,
|
| 1651 |
+
"learning_rate": 4.2119768222931865e-06,
|
| 1652 |
+
"loss": 0.0109588623046875,
|
| 1653 |
+
"step": 1090
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 1.8469000421762969,
|
| 1657 |
+
"grad_norm": 0.036163256401816654,
|
| 1658 |
+
"learning_rate": 4.19742121031913e-06,
|
| 1659 |
+
"loss": 0.012054443359375,
|
| 1660 |
+
"step": 1095
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 1.855335301560523,
|
| 1664 |
+
"grad_norm": 0.04078407955383746,
|
| 1665 |
+
"learning_rate": 4.182831990838709e-06,
|
| 1666 |
+
"loss": 0.0132843017578125,
|
| 1667 |
+
"step": 1100
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 1.855335301560523,
|
| 1671 |
+
"eval_loss": 0.003225065069273114,
|
| 1672 |
+
"eval_margin": -0.008749207222623932,
|
| 1673 |
+
"eval_mean_neg": 0.49084940552711487,
|
| 1674 |
+
"eval_mean_pos": 0.7849159836769104,
|
| 1675 |
+
"eval_runtime": 366.687,
|
| 1676 |
+
"eval_samples_per_second": 21.787,
|
| 1677 |
+
"eval_steps_per_second": 0.341,
|
| 1678 |
+
"step": 1100
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 1.863770560944749,
|
| 1682 |
+
"grad_norm": 0.047827239751426935,
|
| 1683 |
+
"learning_rate": 4.168209573321271e-06,
|
| 1684 |
+
"loss": 0.0133697509765625,
|
| 1685 |
+
"step": 1105
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 1.872205820328975,
|
| 1689 |
+
"grad_norm": 0.0274823880547768,
|
| 1690 |
+
"learning_rate": 4.153554368167927e-06,
|
| 1691 |
+
"loss": 0.010877227783203125,
|
| 1692 |
+
"step": 1110
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 1.8806410797132012,
|
| 1696 |
+
"grad_norm": 0.052787755841206804,
|
| 1697 |
+
"learning_rate": 4.138866786700016e-06,
|
| 1698 |
+
"loss": 0.0139434814453125,
|
| 1699 |
+
"step": 1115
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 1.8890763390974272,
|
| 1703 |
+
"grad_norm": 0.029629846825489692,
|
| 1704 |
+
"learning_rate": 4.124147241147577e-06,
|
| 1705 |
+
"loss": 0.011189651489257813,
|
| 1706 |
+
"step": 1120
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 1.8975115984816533,
|
| 1710 |
+
"grad_norm": 0.039855575258898726,
|
| 1711 |
+
"learning_rate": 4.109396144637764e-06,
|
| 1712 |
+
"loss": 0.010993194580078126,
|
| 1713 |
+
"step": 1125
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 1.9059468578658794,
|
| 1717 |
+
"grad_norm": 0.03789188882991695,
|
| 1718 |
+
"learning_rate": 4.094613911183265e-06,
|
| 1719 |
+
"loss": 0.01313323974609375,
|
| 1720 |
+
"step": 1130
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 1.9143821172501054,
|
| 1724 |
+
"grad_norm": 0.03482605825228896,
|
| 1725 |
+
"learning_rate": 4.0798009556706685e-06,
|
| 1726 |
+
"loss": 0.008492279052734374,
|
| 1727 |
+
"step": 1135
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 1.9228173766343315,
|
| 1731 |
+
"grad_norm": 0.0395626147511318,
|
| 1732 |
+
"learning_rate": 4.064957693848831e-06,
|
| 1733 |
+
"loss": 0.011167144775390625,
|
| 1734 |
+
"step": 1140
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 1.9312526360185576,
|
| 1738 |
+
"grad_norm": 0.026910728579180684,
|
| 1739 |
+
"learning_rate": 4.050084542317201e-06,
|
| 1740 |
+
"loss": 0.0124908447265625,
|
| 1741 |
+
"step": 1145
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 1.9396878954027836,
|
| 1745 |
+
"grad_norm": 0.05111929237613795,
|
| 1746 |
+
"learning_rate": 4.0351819185141284e-06,
|
| 1747 |
+
"loss": 0.01279144287109375,
|
| 1748 |
+
"step": 1150
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 1.9481231547870097,
|
| 1752 |
+
"grad_norm": 0.031631097839140386,
|
| 1753 |
+
"learning_rate": 4.02025024070515e-06,
|
| 1754 |
+
"loss": 0.010783004760742187,
|
| 1755 |
+
"step": 1155
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 1.9565584141712358,
|
| 1759 |
+
"grad_norm": 0.03921591693735718,
|
| 1760 |
+
"learning_rate": 4.005289927971248e-06,
|
| 1761 |
+
"loss": 0.009867095947265625,
|
| 1762 |
+
"step": 1160
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 1.9649936735554618,
|
| 1766 |
+
"grad_norm": 0.03786979993880419,
|
| 1767 |
+
"learning_rate": 3.990301400197088e-06,
|
| 1768 |
+
"loss": 0.010943603515625,
|
| 1769 |
+
"step": 1165
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 1.973428932939688,
|
| 1773 |
+
"grad_norm": 0.033688024912648086,
|
| 1774 |
+
"learning_rate": 3.9752850780592366e-06,
|
| 1775 |
+
"loss": 0.010836410522460937,
|
| 1776 |
+
"step": 1170
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 1.981864192323914,
|
| 1780 |
+
"grad_norm": 0.0473160707405277,
|
| 1781 |
+
"learning_rate": 3.960241383014353e-06,
|
| 1782 |
+
"loss": 0.011658477783203124,
|
| 1783 |
+
"step": 1175
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 1.99029945170814,
|
| 1787 |
+
"grad_norm": 0.034470209590808834,
|
| 1788 |
+
"learning_rate": 3.945170737287356e-06,
|
| 1789 |
+
"loss": 0.0096588134765625,
|
| 1790 |
+
"step": 1180
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 1.9987347110923661,
|
| 1794 |
+
"grad_norm": 0.04035006428036731,
|
| 1795 |
+
"learning_rate": 3.930073563859583e-06,
|
| 1796 |
+
"loss": 0.013312530517578126,
|
| 1797 |
+
"step": 1185
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 2.006748207507381,
|
| 1801 |
+
"grad_norm": 0.03443773853658945,
|
| 1802 |
+
"learning_rate": 3.914950286456911e-06,
|
| 1803 |
+
"loss": 0.0104766845703125,
|
| 1804 |
+
"step": 1190
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 2.015183466891607,
|
| 1808 |
+
"grad_norm": 0.02321269258461312,
|
| 1809 |
+
"learning_rate": 3.899801329537865e-06,
|
| 1810 |
+
"loss": 0.008111572265625,
|
| 1811 |
+
"step": 1195
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 2.023618726275833,
|
| 1815 |
+
"grad_norm": 0.02427731911492366,
|
| 1816 |
+
"learning_rate": 3.884627118281706e-06,
|
| 1817 |
+
"loss": 0.009668731689453125,
|
| 1818 |
+
"step": 1200
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 2.023618726275833,
|
| 1822 |
+
"eval_loss": 0.0028827113565057516,
|
| 1823 |
+
"eval_margin": -0.008073512017877111,
|
| 1824 |
+
"eval_mean_neg": 0.5066258907318115,
|
| 1825 |
+
"eval_mean_pos": 0.7934735417366028,
|
| 1826 |
+
"eval_runtime": 365.1842,
|
| 1827 |
+
"eval_samples_per_second": 21.877,
|
| 1828 |
+
"eval_steps_per_second": 0.342,
|
| 1829 |
+
"step": 1200
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.0320539856600592,
|
| 1833 |
+
"grad_norm": 0.033776934236771874,
|
| 1834 |
+
"learning_rate": 3.869428078576498e-06,
|
| 1835 |
+
"loss": 0.00937347412109375,
|
| 1836 |
+
"step": 1205
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 2.0404892450442853,
|
| 1840 |
+
"grad_norm": 0.05070270762284893,
|
| 1841 |
+
"learning_rate": 3.8542046370071575e-06,
|
| 1842 |
+
"loss": 0.008733367919921875,
|
| 1843 |
+
"step": 1210
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 2.0489245044285114,
|
| 1847 |
+
"grad_norm": 0.028063560546546604,
|
| 1848 |
+
"learning_rate": 3.838957220843472e-06,
|
| 1849 |
+
"loss": 0.00914459228515625,
|
| 1850 |
+
"step": 1215
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.0573597638127374,
|
| 1854 |
+
"grad_norm": 0.041287537117132886,
|
| 1855 |
+
"learning_rate": 3.8236862580281175e-06,
|
| 1856 |
+
"loss": 0.010516357421875,
|
| 1857 |
+
"step": 1220
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.0657950231969635,
|
| 1861 |
+
"grad_norm": 0.03173632436563901,
|
| 1862 |
+
"learning_rate": 3.808392177164642e-06,
|
| 1863 |
+
"loss": 0.010186767578125,
|
| 1864 |
+
"step": 1225
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.0742302825811896,
|
| 1868 |
+
"grad_norm": 0.03149301964970768,
|
| 1869 |
+
"learning_rate": 3.7930754075054406e-06,
|
| 1870 |
+
"loss": 0.010378265380859375,
|
| 1871 |
+
"step": 1230
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.0826655419654156,
|
| 1875 |
+
"grad_norm": 0.03183747792195117,
|
| 1876 |
+
"learning_rate": 3.7777363789397004e-06,
|
| 1877 |
+
"loss": 0.009032630920410156,
|
| 1878 |
+
"step": 1235
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.0911008013496417,
|
| 1882 |
+
"grad_norm": 0.0353065686803631,
|
| 1883 |
+
"learning_rate": 3.7623755219813442e-06,
|
| 1884 |
+
"loss": 0.0096771240234375,
|
| 1885 |
+
"step": 1240
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 2.0995360607338673,
|
| 1889 |
+
"grad_norm": 0.040800577074973816,
|
| 1890 |
+
"learning_rate": 3.746993267756939e-06,
|
| 1891 |
+
"loss": 0.009685516357421875,
|
| 1892 |
+
"step": 1245
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.107971320118094,
|
| 1896 |
+
"grad_norm": 0.04064182954953987,
|
| 1897 |
+
"learning_rate": 3.7315900479936044e-06,
|
| 1898 |
+
"loss": 0.010097503662109375,
|
| 1899 |
+
"step": 1250
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.1164065795023195,
|
| 1903 |
+
"grad_norm": 0.04908593416403285,
|
| 1904 |
+
"learning_rate": 3.7161662950068846e-06,
|
| 1905 |
+
"loss": 0.009412384033203125,
|
| 1906 |
+
"step": 1255
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.124841838886546,
|
| 1910 |
+
"grad_norm": 0.0284060145446946,
|
| 1911 |
+
"learning_rate": 3.7007224416886276e-06,
|
| 1912 |
+
"loss": 0.00821533203125,
|
| 1913 |
+
"step": 1260
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.1332770982707716,
|
| 1917 |
+
"grad_norm": 0.030842726867602113,
|
| 1918 |
+
"learning_rate": 3.685258921494824e-06,
|
| 1919 |
+
"loss": 0.009014129638671875,
|
| 1920 |
+
"step": 1265
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.141712357654998,
|
| 1924 |
+
"grad_norm": 0.036273158990138075,
|
| 1925 |
+
"learning_rate": 3.6697761684334466e-06,
|
| 1926 |
+
"loss": 0.010558700561523438,
|
| 1927 |
+
"step": 1270
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 2.1501476170392237,
|
| 1931 |
+
"grad_norm": 0.03693819496482909,
|
| 1932 |
+
"learning_rate": 3.6542746170522717e-06,
|
| 1933 |
+
"loss": 0.010668182373046875,
|
| 1934 |
+
"step": 1275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.15858287642345,
|
| 1938 |
+
"grad_norm": 0.03797267942950567,
|
| 1939 |
+
"learning_rate": 3.638754702426678e-06,
|
| 1940 |
+
"loss": 0.008889389038085938,
|
| 1941 |
+
"step": 1280
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 2.167018135807676,
|
| 1945 |
+
"grad_norm": 0.03341108305444907,
|
| 1946 |
+
"learning_rate": 3.6232168601474363e-06,
|
| 1947 |
+
"loss": 0.006923675537109375,
|
| 1948 |
+
"step": 1285
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 2.175453395191902,
|
| 1952 |
+
"grad_norm": 0.03454779917085028,
|
| 1953 |
+
"learning_rate": 3.607661526308488e-06,
|
| 1954 |
+
"loss": 0.00969085693359375,
|
| 1955 |
+
"step": 1290
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.183888654576128,
|
| 1959 |
+
"grad_norm": 0.035727285557249105,
|
| 1960 |
+
"learning_rate": 3.5920891374947005e-06,
|
| 1961 |
+
"loss": 0.00997161865234375,
|
| 1962 |
+
"step": 1295
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 2.192323913960354,
|
| 1966 |
+
"grad_norm": 0.031320211315080816,
|
| 1967 |
+
"learning_rate": 3.5765001307696152e-06,
|
| 1968 |
+
"loss": 0.007769393920898438,
|
| 1969 |
+
"step": 1300
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.192323913960354,
|
| 1973 |
+
"eval_loss": 0.0026897923089563847,
|
| 1974 |
+
"eval_margin": -0.0077478337221808975,
|
| 1975 |
+
"eval_mean_neg": 0.49285975098609924,
|
| 1976 |
+
"eval_mean_pos": 0.7862820625305176,
|
| 1977 |
+
"eval_runtime": 362.3777,
|
| 1978 |
+
"eval_samples_per_second": 22.046,
|
| 1979 |
+
"eval_steps_per_second": 0.345,
|
| 1980 |
+
"step": 1300
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 2.20075917334458,
|
| 1984 |
+
"grad_norm": 0.023952498523963275,
|
| 1985 |
+
"learning_rate": 3.560894943663185e-06,
|
| 1986 |
+
"loss": 0.009902191162109376,
|
| 1987 |
+
"step": 1305
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 2.2091944327288062,
|
| 1991 |
+
"grad_norm": 0.029616458459003896,
|
| 1992 |
+
"learning_rate": 3.545274014159486e-06,
|
| 1993 |
+
"loss": 0.008718109130859375,
|
| 1994 |
+
"step": 1310
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 2.2176296921130323,
|
| 1998 |
+
"grad_norm": 0.026768679077660198,
|
| 1999 |
+
"learning_rate": 3.5296377806844334e-06,
|
| 2000 |
+
"loss": 0.006624603271484375,
|
| 2001 |
+
"step": 1315
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 2.2260649514972584,
|
| 2005 |
+
"grad_norm": 0.03723135315427558,
|
| 2006 |
+
"learning_rate": 3.5139866820934687e-06,
|
| 2007 |
+
"loss": 0.010486793518066407,
|
| 2008 |
+
"step": 1320
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 2.2345002108814844,
|
| 2012 |
+
"grad_norm": 0.030973900207479872,
|
| 2013 |
+
"learning_rate": 3.498321157659248e-06,
|
| 2014 |
+
"loss": 0.00841064453125,
|
| 2015 |
+
"step": 1325
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.2429354702657105,
|
| 2019 |
+
"grad_norm": 0.042522927349784224,
|
| 2020 |
+
"learning_rate": 3.482641647059313e-06,
|
| 2021 |
+
"loss": 0.010484886169433594,
|
| 2022 |
+
"step": 1330
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 2.2513707296499366,
|
| 2026 |
+
"grad_norm": 0.036298357689256384,
|
| 2027 |
+
"learning_rate": 3.4669485903637452e-06,
|
| 2028 |
+
"loss": 0.010845947265625,
|
| 2029 |
+
"step": 1335
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 2.2598059890341626,
|
| 2033 |
+
"grad_norm": 0.04210885166855473,
|
| 2034 |
+
"learning_rate": 3.4512424280228227e-06,
|
| 2035 |
+
"loss": 0.009656906127929688,
|
| 2036 |
+
"step": 1340
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 2.2682412484183887,
|
| 2040 |
+
"grad_norm": 0.037852259539673916,
|
| 2041 |
+
"learning_rate": 3.435523600854652e-06,
|
| 2042 |
+
"loss": 0.009561920166015625,
|
| 2043 |
+
"step": 1345
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 2.2766765078026148,
|
| 2047 |
+
"grad_norm": 0.03972030283651443,
|
| 2048 |
+
"learning_rate": 3.4197925500327973e-06,
|
| 2049 |
+
"loss": 0.00974578857421875,
|
| 2050 |
+
"step": 1350
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 2.285111767186841,
|
| 2054 |
+
"grad_norm": 0.03864567979018308,
|
| 2055 |
+
"learning_rate": 3.4040497170739e-06,
|
| 2056 |
+
"loss": 0.009082794189453125,
|
| 2057 |
+
"step": 1355
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 2.293547026571067,
|
| 2061 |
+
"grad_norm": 0.03547766099076331,
|
| 2062 |
+
"learning_rate": 3.3882955438252852e-06,
|
| 2063 |
+
"loss": 0.008104705810546875,
|
| 2064 |
+
"step": 1360
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 2.301982285955293,
|
| 2068 |
+
"grad_norm": 0.042069666240123815,
|
| 2069 |
+
"learning_rate": 3.372530472452561e-06,
|
| 2070 |
+
"loss": 0.010825538635253906,
|
| 2071 |
+
"step": 1365
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 2.310417545339519,
|
| 2075 |
+
"grad_norm": 0.030187240942476403,
|
| 2076 |
+
"learning_rate": 3.356754945427209e-06,
|
| 2077 |
+
"loss": 0.010921478271484375,
|
| 2078 |
+
"step": 1370
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 2.318852804723745,
|
| 2082 |
+
"grad_norm": 0.03775236120881388,
|
| 2083 |
+
"learning_rate": 3.3409694055141636e-06,
|
| 2084 |
+
"loss": 0.00971527099609375,
|
| 2085 |
+
"step": 1375
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 2.327288064107971,
|
| 2089 |
+
"grad_norm": 0.04517333042895106,
|
| 2090 |
+
"learning_rate": 3.3251742957593896e-06,
|
| 2091 |
+
"loss": 0.010394287109375,
|
| 2092 |
+
"step": 1380
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 2.3357233234921972,
|
| 2096 |
+
"grad_norm": 0.03441694727754078,
|
| 2097 |
+
"learning_rate": 3.3093700594774415e-06,
|
| 2098 |
+
"loss": 0.008525848388671875,
|
| 2099 |
+
"step": 1385
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 2.3441585828764233,
|
| 2103 |
+
"grad_norm": 0.039855958024762626,
|
| 2104 |
+
"learning_rate": 3.2935571402390243e-06,
|
| 2105 |
+
"loss": 0.01035003662109375,
|
| 2106 |
+
"step": 1390
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 2.3525938422606494,
|
| 2110 |
+
"grad_norm": 0.036912654679360425,
|
| 2111 |
+
"learning_rate": 3.2777359818585453e-06,
|
| 2112 |
+
"loss": 0.01036224365234375,
|
| 2113 |
+
"step": 1395
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 2.3610291016448754,
|
| 2117 |
+
"grad_norm": 0.02819486898709386,
|
| 2118 |
+
"learning_rate": 3.2619070283816567e-06,
|
| 2119 |
+
"loss": 0.008788299560546876,
|
| 2120 |
+
"step": 1400
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 2.3610291016448754,
|
| 2124 |
+
"eval_loss": 0.002835027640685439,
|
| 2125 |
+
"eval_margin": -0.006975951657119778,
|
| 2126 |
+
"eval_mean_neg": 0.4969240725040436,
|
| 2127 |
+
"eval_mean_pos": 0.7925288081169128,
|
| 2128 |
+
"eval_runtime": 364.0594,
|
| 2129 |
+
"eval_samples_per_second": 21.944,
|
| 2130 |
+
"eval_steps_per_second": 0.343,
|
| 2131 |
+
"step": 1400
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.3694643610291015,
|
| 2135 |
+
"grad_norm": 0.0347736325148637,
|
| 2136 |
+
"learning_rate": 3.24607072407279e-06,
|
| 2137 |
+
"loss": 0.00931854248046875,
|
| 2138 |
+
"step": 1405
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.3778996204133276,
|
| 2142 |
+
"grad_norm": 0.02843547221351205,
|
| 2143 |
+
"learning_rate": 3.2302275134026902e-06,
|
| 2144 |
+
"loss": 0.008514404296875,
|
| 2145 |
+
"step": 1410
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.3863348797975537,
|
| 2149 |
+
"grad_norm": 0.035527939183407756,
|
| 2150 |
+
"learning_rate": 3.2143778410359414e-06,
|
| 2151 |
+
"loss": 0.009189605712890625,
|
| 2152 |
+
"step": 1415
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.3947701391817797,
|
| 2156 |
+
"grad_norm": 0.02697400462877436,
|
| 2157 |
+
"learning_rate": 3.1985221518184845e-06,
|
| 2158 |
+
"loss": 0.008056259155273438,
|
| 2159 |
+
"step": 1420
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.403205398566006,
|
| 2163 |
+
"grad_norm": 0.02974726363919492,
|
| 2164 |
+
"learning_rate": 3.1826608907651327e-06,
|
| 2165 |
+
"loss": 0.008675384521484374,
|
| 2166 |
+
"step": 1425
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.411640657950232,
|
| 2170 |
+
"grad_norm": 0.04279127831329293,
|
| 2171 |
+
"learning_rate": 3.1667945030470815e-06,
|
| 2172 |
+
"loss": 0.009341812133789063,
|
| 2173 |
+
"step": 1430
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.420075917334458,
|
| 2177 |
+
"grad_norm": 0.039837807919925805,
|
| 2178 |
+
"learning_rate": 3.1509234339794144e-06,
|
| 2179 |
+
"loss": 0.010208892822265624,
|
| 2180 |
+
"step": 1435
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.428511176718684,
|
| 2184 |
+
"grad_norm": 0.03024657864136027,
|
| 2185 |
+
"learning_rate": 3.1350481290086038e-06,
|
| 2186 |
+
"loss": 0.008173370361328125,
|
| 2187 |
+
"step": 1440
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.43694643610291,
|
| 2191 |
+
"grad_norm": 0.03564318900525913,
|
| 2192 |
+
"learning_rate": 3.119169033700011e-06,
|
| 2193 |
+
"loss": 0.00924224853515625,
|
| 2194 |
+
"step": 1445
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.445381695487136,
|
| 2198 |
+
"grad_norm": 0.028913985964356455,
|
| 2199 |
+
"learning_rate": 3.103286593725377e-06,
|
| 2200 |
+
"loss": 0.008563995361328125,
|
| 2201 |
+
"step": 1450
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.453816954871362,
|
| 2205 |
+
"grad_norm": 0.035875161756803144,
|
| 2206 |
+
"learning_rate": 3.0874012548503173e-06,
|
| 2207 |
+
"loss": 0.009112548828125,
|
| 2208 |
+
"step": 1455
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.4622522142555883,
|
| 2212 |
+
"grad_norm": 0.03817913502015442,
|
| 2213 |
+
"learning_rate": 3.0715134629218095e-06,
|
| 2214 |
+
"loss": 0.007489013671875,
|
| 2215 |
+
"step": 1460
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.4706874736398143,
|
| 2219 |
+
"grad_norm": 0.03470677728941542,
|
| 2220 |
+
"learning_rate": 3.0556236638556803e-06,
|
| 2221 |
+
"loss": 0.012370681762695313,
|
| 2222 |
+
"step": 1465
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.4791227330240404,
|
| 2226 |
+
"grad_norm": 0.042966141209856486,
|
| 2227 |
+
"learning_rate": 3.0397323036240886e-06,
|
| 2228 |
+
"loss": 0.0088165283203125,
|
| 2229 |
+
"step": 1470
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.4875579924082665,
|
| 2233 |
+
"grad_norm": 0.03434953324492014,
|
| 2234 |
+
"learning_rate": 3.023839828243012e-06,
|
| 2235 |
+
"loss": 0.008261871337890626,
|
| 2236 |
+
"step": 1475
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.4959932517924925,
|
| 2240 |
+
"grad_norm": 0.03061507966476803,
|
| 2241 |
+
"learning_rate": 3.007946683759723e-06,
|
| 2242 |
+
"loss": 0.008873748779296874,
|
| 2243 |
+
"step": 1480
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.5044285111767186,
|
| 2247 |
+
"grad_norm": 0.027732115444419583,
|
| 2248 |
+
"learning_rate": 2.9920533162402776e-06,
|
| 2249 |
+
"loss": 0.008371734619140625,
|
| 2250 |
+
"step": 1485
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.5128637705609447,
|
| 2254 |
+
"grad_norm": 0.029951392389848317,
|
| 2255 |
+
"learning_rate": 2.9761601717569896e-06,
|
| 2256 |
+
"loss": 0.00865478515625,
|
| 2257 |
+
"step": 1490
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.5212990299451707,
|
| 2261 |
+
"grad_norm": 0.033652436341082566,
|
| 2262 |
+
"learning_rate": 2.960267696375911e-06,
|
| 2263 |
+
"loss": 0.009691619873046875,
|
| 2264 |
+
"step": 1495
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.529734289329397,
|
| 2268 |
+
"grad_norm": 0.024511774862390433,
|
| 2269 |
+
"learning_rate": 2.9443763361443203e-06,
|
| 2270 |
+
"loss": 0.010028076171875,
|
| 2271 |
+
"step": 1500
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.529734289329397,
|
| 2275 |
+
"eval_loss": 0.002559108193963766,
|
| 2276 |
+
"eval_margin": -0.006972289358776423,
|
| 2277 |
+
"eval_mean_neg": 0.4843982458114624,
|
| 2278 |
+
"eval_mean_pos": 0.7881345152854919,
|
| 2279 |
+
"eval_runtime": 365.5243,
|
| 2280 |
+
"eval_samples_per_second": 21.856,
|
| 2281 |
+
"eval_steps_per_second": 0.342,
|
| 2282 |
+
"step": 1500
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 2.538169548713623,
|
| 2286 |
+
"grad_norm": 0.034627287332041165,
|
| 2287 |
+
"learning_rate": 2.9284865370781906e-06,
|
| 2288 |
+
"loss": 0.00982513427734375,
|
| 2289 |
+
"step": 1505
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 2.546604808097849,
|
| 2293 |
+
"grad_norm": 0.03482839500691478,
|
| 2294 |
+
"learning_rate": 2.9125987451496837e-06,
|
| 2295 |
+
"loss": 0.00842742919921875,
|
| 2296 |
+
"step": 1510
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 2.555040067482075,
|
| 2300 |
+
"grad_norm": 0.034040276652772095,
|
| 2301 |
+
"learning_rate": 2.8967134062746236e-06,
|
| 2302 |
+
"loss": 0.008990859985351563,
|
| 2303 |
+
"step": 1515
|
| 2304 |
+
},
|
| 2305 |
+
{
|
| 2306 |
+
"epoch": 2.563475326866301,
|
| 2307 |
+
"grad_norm": 0.03868034786852329,
|
| 2308 |
+
"learning_rate": 2.8808309662999897e-06,
|
| 2309 |
+
"loss": 0.007648468017578125,
|
| 2310 |
+
"step": 1520
|
| 2311 |
+
},
|
| 2312 |
+
{
|
| 2313 |
+
"epoch": 2.571910586250527,
|
| 2314 |
+
"grad_norm": 0.0419385930164125,
|
| 2315 |
+
"learning_rate": 2.864951870991397e-06,
|
| 2316 |
+
"loss": 0.009268951416015626,
|
| 2317 |
+
"step": 1525
|
| 2318 |
+
},
|
| 2319 |
+
{
|
| 2320 |
+
"epoch": 2.580345845634753,
|
| 2321 |
+
"grad_norm": 0.03690945718603307,
|
| 2322 |
+
"learning_rate": 2.8490765660205857e-06,
|
| 2323 |
+
"loss": 0.00864715576171875,
|
| 2324 |
+
"step": 1530
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"epoch": 2.5887811050189793,
|
| 2328 |
+
"grad_norm": 0.02493335874585237,
|
| 2329 |
+
"learning_rate": 2.833205496952919e-06,
|
| 2330 |
+
"loss": 0.00865478515625,
|
| 2331 |
+
"step": 1535
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 2.5972163644032054,
|
| 2335 |
+
"grad_norm": 0.0355467734297459,
|
| 2336 |
+
"learning_rate": 2.817339109234868e-06,
|
| 2337 |
+
"loss": 0.009038543701171875,
|
| 2338 |
+
"step": 1540
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 2.6056516237874314,
|
| 2342 |
+
"grad_norm": 0.035706551369837684,
|
| 2343 |
+
"learning_rate": 2.801477848181517e-06,
|
| 2344 |
+
"loss": 0.008769607543945313,
|
| 2345 |
+
"step": 1545
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 2.6140868831716575,
|
| 2349 |
+
"grad_norm": 0.030590948482880534,
|
| 2350 |
+
"learning_rate": 2.7856221589640584e-06,
|
| 2351 |
+
"loss": 0.010419464111328125,
|
| 2352 |
+
"step": 1550
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 2.6225221425558836,
|
| 2356 |
+
"grad_norm": 0.031187166990055255,
|
| 2357 |
+
"learning_rate": 2.7697724865973103e-06,
|
| 2358 |
+
"loss": 0.008966064453125,
|
| 2359 |
+
"step": 1555
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 2.6309574019401096,
|
| 2363 |
+
"grad_norm": 0.03195446103788609,
|
| 2364 |
+
"learning_rate": 2.753929275927211e-06,
|
| 2365 |
+
"loss": 0.00810089111328125,
|
| 2366 |
+
"step": 1560
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 2.6393926613243357,
|
| 2370 |
+
"grad_norm": 0.03124766219549955,
|
| 2371 |
+
"learning_rate": 2.7380929716183448e-06,
|
| 2372 |
+
"loss": 0.00867919921875,
|
| 2373 |
+
"step": 1565
|
| 2374 |
+
},
|
| 2375 |
+
{
|
| 2376 |
+
"epoch": 2.6478279207085618,
|
| 2377 |
+
"grad_norm": 0.04158743972175772,
|
| 2378 |
+
"learning_rate": 2.722264018141455e-06,
|
| 2379 |
+
"loss": 0.008811187744140626,
|
| 2380 |
+
"step": 1570
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 2.656263180092788,
|
| 2384 |
+
"grad_norm": 0.042358151513616535,
|
| 2385 |
+
"learning_rate": 2.706442859760976e-06,
|
| 2386 |
+
"loss": 0.008480644226074219,
|
| 2387 |
+
"step": 1575
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 2.664698439477014,
|
| 2391 |
+
"grad_norm": 0.02876853915749735,
|
| 2392 |
+
"learning_rate": 2.6906299405225595e-06,
|
| 2393 |
+
"loss": 0.009603309631347656,
|
| 2394 |
+
"step": 1580
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 2.67313369886124,
|
| 2398 |
+
"grad_norm": 0.031452133973887623,
|
| 2399 |
+
"learning_rate": 2.6748257042406114e-06,
|
| 2400 |
+
"loss": 0.008524322509765625,
|
| 2401 |
+
"step": 1585
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 2.681568958245466,
|
| 2405 |
+
"grad_norm": 0.03502577600676223,
|
| 2406 |
+
"learning_rate": 2.659030594485836e-06,
|
| 2407 |
+
"loss": 0.007845306396484375,
|
| 2408 |
+
"step": 1590
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 2.690004217629692,
|
| 2412 |
+
"grad_norm": 0.029358280910616305,
|
| 2413 |
+
"learning_rate": 2.6432450545727913e-06,
|
| 2414 |
+
"loss": 0.008304595947265625,
|
| 2415 |
+
"step": 1595
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 2.698439477013918,
|
| 2419 |
+
"grad_norm": 0.037226468621806945,
|
| 2420 |
+
"learning_rate": 2.62746952754744e-06,
|
| 2421 |
+
"loss": 0.0089141845703125,
|
| 2422 |
+
"step": 1600
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 2.698439477013918,
|
| 2426 |
+
"eval_loss": 0.002468662802129984,
|
| 2427 |
+
"eval_margin": -0.006652700444383006,
|
| 2428 |
+
"eval_mean_neg": 0.5055871605873108,
|
| 2429 |
+
"eval_mean_pos": 0.8004181981086731,
|
| 2430 |
+
"eval_runtime": 363.3432,
|
| 2431 |
+
"eval_samples_per_second": 21.987,
|
| 2432 |
+
"eval_steps_per_second": 0.344,
|
| 2433 |
+
"step": 1600
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.7068747363981442,
|
| 2437 |
+
"grad_norm": 0.029979441347867175,
|
| 2438 |
+
"learning_rate": 2.6117044561747145e-06,
|
| 2439 |
+
"loss": 0.007899856567382813,
|
| 2440 |
+
"step": 1605
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.7153099957823703,
|
| 2444 |
+
"grad_norm": 0.04117264280378634,
|
| 2445 |
+
"learning_rate": 2.5959502829261e-06,
|
| 2446 |
+
"loss": 0.009801483154296875,
|
| 2447 |
+
"step": 1610
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.7237452551665964,
|
| 2451 |
+
"grad_norm": 0.02874139529420723,
|
| 2452 |
+
"learning_rate": 2.5802074499672033e-06,
|
| 2453 |
+
"loss": 0.007126617431640625,
|
| 2454 |
+
"step": 1615
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 2.7321805145508224,
|
| 2458 |
+
"grad_norm": 0.032009387593884574,
|
| 2459 |
+
"learning_rate": 2.564476399145349e-06,
|
| 2460 |
+
"loss": 0.007319259643554688,
|
| 2461 |
+
"step": 1620
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 2.7406157739350485,
|
| 2465 |
+
"grad_norm": 0.0343660828009257,
|
| 2466 |
+
"learning_rate": 2.5487575719771774e-06,
|
| 2467 |
+
"loss": 0.010648345947265625,
|
| 2468 |
+
"step": 1625
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 2.7490510333192746,
|
| 2472 |
+
"grad_norm": 0.033859872264591424,
|
| 2473 |
+
"learning_rate": 2.533051409636255e-06,
|
| 2474 |
+
"loss": 0.007244110107421875,
|
| 2475 |
+
"step": 1630
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 2.7574862927035007,
|
| 2479 |
+
"grad_norm": 0.032973506044290384,
|
| 2480 |
+
"learning_rate": 2.517358352940688e-06,
|
| 2481 |
+
"loss": 0.008284759521484376,
|
| 2482 |
+
"step": 1635
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 2.7659215520877267,
|
| 2486 |
+
"grad_norm": 0.03481146191160576,
|
| 2487 |
+
"learning_rate": 2.501678842340753e-06,
|
| 2488 |
+
"loss": 0.00882110595703125,
|
| 2489 |
+
"step": 1640
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.774356811471953,
|
| 2493 |
+
"grad_norm": 0.03862588539253724,
|
| 2494 |
+
"learning_rate": 2.4860133179065323e-06,
|
| 2495 |
+
"loss": 0.00964202880859375,
|
| 2496 |
+
"step": 1645
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 2.782792070856179,
|
| 2500 |
+
"grad_norm": 0.02979780702601001,
|
| 2501 |
+
"learning_rate": 2.4703622193155676e-06,
|
| 2502 |
+
"loss": 0.009095001220703124,
|
| 2503 |
+
"step": 1650
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 2.791227330240405,
|
| 2507 |
+
"grad_norm": 0.02658002258647219,
|
| 2508 |
+
"learning_rate": 2.4547259858405147e-06,
|
| 2509 |
+
"loss": 0.008580398559570313,
|
| 2510 |
+
"step": 1655
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.799662589624631,
|
| 2514 |
+
"grad_norm": 0.03237100489547251,
|
| 2515 |
+
"learning_rate": 2.439105056336816e-06,
|
| 2516 |
+
"loss": 0.006137275695800781,
|
| 2517 |
+
"step": 1660
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 2.808097849008857,
|
| 2521 |
+
"grad_norm": 0.035925961611001624,
|
| 2522 |
+
"learning_rate": 2.423499869230385e-06,
|
| 2523 |
+
"loss": 0.006979179382324219,
|
| 2524 |
+
"step": 1665
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 2.816533108393083,
|
| 2528 |
+
"grad_norm": 0.028925897672990208,
|
| 2529 |
+
"learning_rate": 2.4079108625053e-06,
|
| 2530 |
+
"loss": 0.007439422607421875,
|
| 2531 |
+
"step": 1670
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 2.824968367777309,
|
| 2535 |
+
"grad_norm": 0.02643424196739614,
|
| 2536 |
+
"learning_rate": 2.392338473691513e-06,
|
| 2537 |
+
"loss": 0.007563400268554688,
|
| 2538 |
+
"step": 1675
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 2.8334036271615353,
|
| 2542 |
+
"grad_norm": 0.029469931037551172,
|
| 2543 |
+
"learning_rate": 2.376783139852564e-06,
|
| 2544 |
+
"loss": 0.00782928466796875,
|
| 2545 |
+
"step": 1680
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 2.8418388865457613,
|
| 2549 |
+
"grad_norm": 0.03519097117769341,
|
| 2550 |
+
"learning_rate": 2.3612452975733225e-06,
|
| 2551 |
+
"loss": 0.0081695556640625,
|
| 2552 |
+
"step": 1685
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.8502741459299874,
|
| 2556 |
+
"grad_norm": 0.041842720836538394,
|
| 2557 |
+
"learning_rate": 2.3457253829477284e-06,
|
| 2558 |
+
"loss": 0.00938720703125,
|
| 2559 |
+
"step": 1690
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 2.8587094053142135,
|
| 2563 |
+
"grad_norm": 0.02803118980318521,
|
| 2564 |
+
"learning_rate": 2.3302238315665544e-06,
|
| 2565 |
+
"loss": 0.007602310180664063,
|
| 2566 |
+
"step": 1695
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 2.8671446646984395,
|
| 2570 |
+
"grad_norm": 0.06643247372472408,
|
| 2571 |
+
"learning_rate": 2.314741078505177e-06,
|
| 2572 |
+
"loss": 0.009275436401367188,
|
| 2573 |
+
"step": 1700
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 2.8671446646984395,
|
| 2577 |
+
"eval_loss": 0.002381447935476899,
|
| 2578 |
+
"eval_margin": -0.0063614378337778395,
|
| 2579 |
+
"eval_mean_neg": 0.4982295334339142,
|
| 2580 |
+
"eval_mean_pos": 0.7957465648651123,
|
| 2581 |
+
"eval_runtime": 364.3957,
|
| 2582 |
+
"eval_samples_per_second": 21.924,
|
| 2583 |
+
"eval_steps_per_second": 0.343,
|
| 2584 |
+
"step": 1700
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 2.8755799240826656,
|
| 2588 |
+
"grad_norm": 0.04573493937998368,
|
| 2589 |
+
"learning_rate": 2.299277558311373e-06,
|
| 2590 |
+
"loss": 0.008275604248046875,
|
| 2591 |
+
"step": 1705
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 2.8840151834668917,
|
| 2595 |
+
"grad_norm": 0.030855319414577996,
|
| 2596 |
+
"learning_rate": 2.283833704993116e-06,
|
| 2597 |
+
"loss": 0.008497047424316406,
|
| 2598 |
+
"step": 1710
|
| 2599 |
+
},
|
| 2600 |
+
{
|
| 2601 |
+
"epoch": 2.8924504428511177,
|
| 2602 |
+
"grad_norm": 0.03287831217925721,
|
| 2603 |
+
"learning_rate": 2.268409952006397e-06,
|
| 2604 |
+
"loss": 0.006939697265625,
|
| 2605 |
+
"step": 1715
|
| 2606 |
+
},
|
| 2607 |
+
{
|
| 2608 |
+
"epoch": 2.900885702235344,
|
| 2609 |
+
"grad_norm": 0.03738971418410914,
|
| 2610 |
+
"learning_rate": 2.253006732243061e-06,
|
| 2611 |
+
"loss": 0.00982208251953125,
|
| 2612 |
+
"step": 1720
|
| 2613 |
+
},
|
| 2614 |
+
{
|
| 2615 |
+
"epoch": 2.90932096161957,
|
| 2616 |
+
"grad_norm": 0.02295281003302144,
|
| 2617 |
+
"learning_rate": 2.237624478018656e-06,
|
| 2618 |
+
"loss": 0.00743560791015625,
|
| 2619 |
+
"step": 1725
|
| 2620 |
+
},
|
| 2621 |
+
{
|
| 2622 |
+
"epoch": 2.917756221003796,
|
| 2623 |
+
"grad_norm": 0.03960242549923526,
|
| 2624 |
+
"learning_rate": 2.2222636210603002e-06,
|
| 2625 |
+
"loss": 0.008847427368164063,
|
| 2626 |
+
"step": 1730
|
| 2627 |
+
},
|
| 2628 |
+
{
|
| 2629 |
+
"epoch": 2.926191480388022,
|
| 2630 |
+
"grad_norm": 0.04741800625952587,
|
| 2631 |
+
"learning_rate": 2.2069245924945604e-06,
|
| 2632 |
+
"loss": 0.009384918212890624,
|
| 2633 |
+
"step": 1735
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 2.934626739772248,
|
| 2637 |
+
"grad_norm": 0.03409532340357435,
|
| 2638 |
+
"learning_rate": 2.191607822835357e-06,
|
| 2639 |
+
"loss": 0.0076019287109375,
|
| 2640 |
+
"step": 1740
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 2.943061999156474,
|
| 2644 |
+
"grad_norm": 0.0239211291682541,
|
| 2645 |
+
"learning_rate": 2.1763137419718826e-06,
|
| 2646 |
+
"loss": 0.007954025268554687,
|
| 2647 |
+
"step": 1745
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"epoch": 2.9514972585407,
|
| 2651 |
+
"grad_norm": 0.0255275562880085,
|
| 2652 |
+
"learning_rate": 2.161042779156529e-06,
|
| 2653 |
+
"loss": 0.007129669189453125,
|
| 2654 |
+
"step": 1750
|
| 2655 |
+
},
|
| 2656 |
+
{
|
| 2657 |
+
"epoch": 2.9599325179249263,
|
| 2658 |
+
"grad_norm": 0.026100931475016395,
|
| 2659 |
+
"learning_rate": 2.1457953629928426e-06,
|
| 2660 |
+
"loss": 0.007111358642578125,
|
| 2661 |
+
"step": 1755
|
| 2662 |
+
},
|
| 2663 |
+
{
|
| 2664 |
+
"epoch": 2.9683677773091524,
|
| 2665 |
+
"grad_norm": 0.03040565516608014,
|
| 2666 |
+
"learning_rate": 2.1305719214235017e-06,
|
| 2667 |
+
"loss": 0.00856170654296875,
|
| 2668 |
+
"step": 1760
|
| 2669 |
+
},
|
| 2670 |
+
{
|
| 2671 |
+
"epoch": 2.9768030366933784,
|
| 2672 |
+
"grad_norm": 0.031725391054917944,
|
| 2673 |
+
"learning_rate": 2.115372881718295e-06,
|
| 2674 |
+
"loss": 0.00930938720703125,
|
| 2675 |
+
"step": 1765
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 2.9852382960776045,
|
| 2679 |
+
"grad_norm": 0.025864373534585865,
|
| 2680 |
+
"learning_rate": 2.100198670462137e-06,
|
| 2681 |
+
"loss": 0.007320022583007813,
|
| 2682 |
+
"step": 1770
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 2.9936735554618306,
|
| 2686 |
+
"grad_norm": 0.013680490985647303,
|
| 2687 |
+
"learning_rate": 2.0850497135430897e-06,
|
| 2688 |
+
"loss": 0.007777786254882813,
|
| 2689 |
+
"step": 1775
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 3.001687051876845,
|
| 2693 |
+
"grad_norm": 0.03143671946142631,
|
| 2694 |
+
"learning_rate": 2.0699264361404174e-06,
|
| 2695 |
+
"loss": 0.008609771728515625,
|
| 2696 |
+
"step": 1780
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 3.010122311261071,
|
| 2700 |
+
"grad_norm": 0.021237532660395856,
|
| 2701 |
+
"learning_rate": 2.054829262712645e-06,
|
| 2702 |
+
"loss": 0.007422637939453125,
|
| 2703 |
+
"step": 1785
|
| 2704 |
+
},
|
| 2705 |
+
{
|
| 2706 |
+
"epoch": 3.018557570645297,
|
| 2707 |
+
"grad_norm": 0.02970629169587053,
|
| 2708 |
+
"learning_rate": 2.0397586169856488e-06,
|
| 2709 |
+
"loss": 0.008047866821289062,
|
| 2710 |
+
"step": 1790
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 3.0269928300295232,
|
| 2714 |
+
"grad_norm": 0.0280079357370666,
|
| 2715 |
+
"learning_rate": 2.024714921940763e-06,
|
| 2716 |
+
"loss": 0.008725738525390625,
|
| 2717 |
+
"step": 1795
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 3.0354280894137493,
|
| 2721 |
+
"grad_norm": 0.05178206206651836,
|
| 2722 |
+
"learning_rate": 2.0096985998029124e-06,
|
| 2723 |
+
"loss": 0.007384490966796875,
|
| 2724 |
+
"step": 1800
|
| 2725 |
+
},
|
| 2726 |
+
{
|
| 2727 |
+
"epoch": 3.0354280894137493,
|
| 2728 |
+
"eval_loss": 0.0022954940795898438,
|
| 2729 |
+
"eval_margin": -0.005535545939159009,
|
| 2730 |
+
"eval_mean_neg": 0.49804064631462097,
|
| 2731 |
+
"eval_mean_pos": 0.7978142499923706,
|
| 2732 |
+
"eval_runtime": 362.5441,
|
| 2733 |
+
"eval_samples_per_second": 22.036,
|
| 2734 |
+
"eval_steps_per_second": 0.345,
|
| 2735 |
+
"step": 1800
|
| 2736 |
+
}
|
| 2737 |
+
],
|
| 2738 |
+
"logging_steps": 5,
|
| 2739 |
+
"max_steps": 2965,
|
| 2740 |
+
"num_input_tokens_seen": 0,
|
| 2741 |
+
"num_train_epochs": 5,
|
| 2742 |
+
"save_steps": 100,
|
| 2743 |
+
"stateful_callbacks": {
|
| 2744 |
+
"TrainerControl": {
|
| 2745 |
+
"args": {
|
| 2746 |
+
"should_epoch_stop": false,
|
| 2747 |
+
"should_evaluate": false,
|
| 2748 |
+
"should_log": false,
|
| 2749 |
+
"should_save": true,
|
| 2750 |
+
"should_training_stop": false
|
| 2751 |
+
},
|
| 2752 |
+
"attributes": {}
|
| 2753 |
+
}
|
| 2754 |
+
},
|
| 2755 |
+
"total_flos": 2008625567629312.0,
|
| 2756 |
+
"train_batch_size": 4,
|
| 2757 |
+
"trial_name": null,
|
| 2758 |
+
"trial_params": null
|
| 2759 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e67a98968389a98ad4e22d001eae6645ff2ad0cb75b9e2316afcc7dcb356ed91
|
| 3 |
+
size 8888
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zero_to_fp32.py
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
|
| 3 |
+
# Copyright (c) Microsoft Corporation.
|
| 4 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 5 |
+
|
| 6 |
+
# DeepSpeed Team
|
| 7 |
+
|
| 8 |
+
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
| 9 |
+
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
| 10 |
+
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
| 11 |
+
# application.
|
| 12 |
+
#
|
| 13 |
+
# example:
|
| 14 |
+
# python zero_to_fp32.py . output_dir/
|
| 15 |
+
# or
|
| 16 |
+
# python zero_to_fp32.py . output_dir/ --safe_serialization
|
| 17 |
+
|
| 18 |
+
import argparse
|
| 19 |
+
import torch
|
| 20 |
+
import glob
|
| 21 |
+
import math
|
| 22 |
+
import os
|
| 23 |
+
import re
|
| 24 |
+
import gc
|
| 25 |
+
import json
|
| 26 |
+
import numpy as np
|
| 27 |
+
from tqdm import tqdm
|
| 28 |
+
from collections import OrderedDict
|
| 29 |
+
from dataclasses import dataclass
|
| 30 |
+
|
| 31 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 32 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 33 |
+
from deepspeed.utils import logger
|
| 34 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 35 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 36 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
@dataclass
|
| 40 |
+
class zero_model_state:
|
| 41 |
+
buffers: dict()
|
| 42 |
+
param_shapes: dict()
|
| 43 |
+
shared_params: list
|
| 44 |
+
ds_version: int
|
| 45 |
+
frozen_param_shapes: dict()
|
| 46 |
+
frozen_param_fragments: dict()
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
debug = 0
|
| 50 |
+
|
| 51 |
+
# load to cpu
|
| 52 |
+
device = torch.device('cpu')
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
def atoi(text):
|
| 56 |
+
return int(text) if text.isdigit() else text
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def natural_keys(text):
|
| 60 |
+
'''
|
| 61 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 62 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 63 |
+
(See Toothy's implementation in the comments)
|
| 64 |
+
'''
|
| 65 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 69 |
+
if not os.path.isdir(checkpoint_dir):
|
| 70 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 71 |
+
|
| 72 |
+
# there should be only one file
|
| 73 |
+
if zero_stage <= 2:
|
| 74 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 75 |
+
elif zero_stage == 3:
|
| 76 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 77 |
+
|
| 78 |
+
if not os.path.exists(file):
|
| 79 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 80 |
+
|
| 81 |
+
return file
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 85 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 86 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 87 |
+
|
| 88 |
+
if len(ckpt_files) == 0:
|
| 89 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 90 |
+
|
| 91 |
+
return ckpt_files
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
def get_optim_files(checkpoint_dir):
|
| 95 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def get_model_state_files(checkpoint_dir):
|
| 99 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
def parse_model_states(files):
|
| 103 |
+
zero_model_states = []
|
| 104 |
+
for file in files:
|
| 105 |
+
state_dict = torch.load(file, map_location=device, weights_only=False)
|
| 106 |
+
|
| 107 |
+
if BUFFER_NAMES not in state_dict:
|
| 108 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 109 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 110 |
+
if debug:
|
| 111 |
+
print("Found buffers:", buffer_names)
|
| 112 |
+
|
| 113 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 114 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 115 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 116 |
+
|
| 117 |
+
# collect parameters that are included in param_shapes
|
| 118 |
+
param_names = []
|
| 119 |
+
for s in param_shapes:
|
| 120 |
+
for name in s.keys():
|
| 121 |
+
param_names.append(name)
|
| 122 |
+
|
| 123 |
+
# update with frozen parameters
|
| 124 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 125 |
+
if frozen_param_shapes is not None:
|
| 126 |
+
if debug:
|
| 127 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 128 |
+
param_names += list(frozen_param_shapes.keys())
|
| 129 |
+
|
| 130 |
+
# handle shared params
|
| 131 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 132 |
+
|
| 133 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 134 |
+
|
| 135 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 136 |
+
|
| 137 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 138 |
+
param_shapes=param_shapes,
|
| 139 |
+
shared_params=shared_params,
|
| 140 |
+
ds_version=ds_version,
|
| 141 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 142 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 143 |
+
zero_model_states.append(z_model_state)
|
| 144 |
+
|
| 145 |
+
return zero_model_states
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 149 |
+
total_files = len(files)
|
| 150 |
+
state_dicts = []
|
| 151 |
+
for f in tqdm(files, desc='Loading checkpoint shards'):
|
| 152 |
+
state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
|
| 153 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 154 |
+
# and also handle the case where it was already removed by another helper script
|
| 155 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 156 |
+
state_dicts.append(state_dict)
|
| 157 |
+
|
| 158 |
+
if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 159 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 160 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 161 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 162 |
+
|
| 163 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 164 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 165 |
+
# use the max of the partition_count to get the dp world_size.
|
| 166 |
+
|
| 167 |
+
if type(world_size) is list:
|
| 168 |
+
world_size = max(world_size)
|
| 169 |
+
|
| 170 |
+
if world_size != total_files:
|
| 171 |
+
raise ValueError(
|
| 172 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 173 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 174 |
+
)
|
| 175 |
+
|
| 176 |
+
# the groups are named differently in each stage
|
| 177 |
+
if zero_stage <= 2:
|
| 178 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 179 |
+
elif zero_stage == 3:
|
| 180 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 181 |
+
else:
|
| 182 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 183 |
+
|
| 184 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 185 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 186 |
+
|
| 187 |
+
|
| 188 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
|
| 189 |
+
"""
|
| 190 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 191 |
+
|
| 192 |
+
Args:
|
| 193 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 194 |
+
|
| 195 |
+
"""
|
| 196 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 197 |
+
|
| 198 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 199 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 200 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 201 |
+
|
| 202 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 203 |
+
|
| 204 |
+
zero_model_states = parse_model_states(model_files)
|
| 205 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 206 |
+
|
| 207 |
+
if zero_stage <= 2:
|
| 208 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 209 |
+
exclude_frozen_parameters)
|
| 210 |
+
elif zero_stage == 3:
|
| 211 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 212 |
+
exclude_frozen_parameters)
|
| 213 |
+
|
| 214 |
+
|
| 215 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 216 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 217 |
+
return
|
| 218 |
+
|
| 219 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 220 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 221 |
+
|
| 222 |
+
if debug:
|
| 223 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 224 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 225 |
+
|
| 226 |
+
wanted_params = len(frozen_param_shapes)
|
| 227 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 229 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 230 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 231 |
+
|
| 232 |
+
total_params = 0
|
| 233 |
+
total_numel = 0
|
| 234 |
+
for name, shape in frozen_param_shapes.items():
|
| 235 |
+
total_params += 1
|
| 236 |
+
unpartitioned_numel = shape.numel()
|
| 237 |
+
total_numel += unpartitioned_numel
|
| 238 |
+
|
| 239 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 240 |
+
|
| 241 |
+
if debug:
|
| 242 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 243 |
+
|
| 244 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 245 |
+
|
| 246 |
+
|
| 247 |
+
def _has_callable(obj, fn):
|
| 248 |
+
attr = getattr(obj, fn, None)
|
| 249 |
+
return callable(attr)
|
| 250 |
+
|
| 251 |
+
|
| 252 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 253 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 254 |
+
|
| 255 |
+
# Reconstruction protocol:
|
| 256 |
+
#
|
| 257 |
+
# XXX: document this
|
| 258 |
+
|
| 259 |
+
if debug:
|
| 260 |
+
for i in range(world_size):
|
| 261 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 262 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 263 |
+
|
| 264 |
+
# XXX: memory usage doubles here (zero2)
|
| 265 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 266 |
+
merged_single_partition_of_fp32_groups = []
|
| 267 |
+
for i in range(num_param_groups):
|
| 268 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 269 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 270 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 271 |
+
avail_numel = sum(
|
| 272 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 273 |
+
|
| 274 |
+
if debug:
|
| 275 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 276 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 277 |
+
# not asserting if there is a mismatch due to possible padding
|
| 278 |
+
print(f"Have {avail_numel} numels to process.")
|
| 279 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 280 |
+
|
| 281 |
+
# params
|
| 282 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 283 |
+
# out-of-core computing solution
|
| 284 |
+
total_numel = 0
|
| 285 |
+
total_params = 0
|
| 286 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 287 |
+
offset = 0
|
| 288 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 289 |
+
for name, shape in shapes.items():
|
| 290 |
+
|
| 291 |
+
unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
|
| 292 |
+
total_numel += unpartitioned_numel
|
| 293 |
+
total_params += 1
|
| 294 |
+
|
| 295 |
+
if debug:
|
| 296 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 297 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 298 |
+
offset += unpartitioned_numel
|
| 299 |
+
|
| 300 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 301 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 302 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 303 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 304 |
+
align_to = 2 * world_size
|
| 305 |
+
|
| 306 |
+
def zero2_align(x):
|
| 307 |
+
return align_to * math.ceil(x / align_to)
|
| 308 |
+
|
| 309 |
+
if debug:
|
| 310 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 311 |
+
|
| 312 |
+
offset = zero2_align(offset)
|
| 313 |
+
avail_numel = zero2_align(avail_numel)
|
| 314 |
+
|
| 315 |
+
if debug:
|
| 316 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 317 |
+
|
| 318 |
+
# Sanity check
|
| 319 |
+
if offset != avail_numel:
|
| 320 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 321 |
+
|
| 322 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 323 |
+
|
| 324 |
+
|
| 325 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 326 |
+
exclude_frozen_parameters):
|
| 327 |
+
state_dict = OrderedDict()
|
| 328 |
+
|
| 329 |
+
# buffers
|
| 330 |
+
buffers = zero_model_states[0].buffers
|
| 331 |
+
state_dict.update(buffers)
|
| 332 |
+
if debug:
|
| 333 |
+
print(f"added {len(buffers)} buffers")
|
| 334 |
+
|
| 335 |
+
if not exclude_frozen_parameters:
|
| 336 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 337 |
+
|
| 338 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 339 |
+
|
| 340 |
+
# recover shared parameters
|
| 341 |
+
for pair in zero_model_states[0].shared_params:
|
| 342 |
+
if pair[1] in state_dict:
|
| 343 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 344 |
+
|
| 345 |
+
return state_dict
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 349 |
+
remainder = unpartitioned_numel % world_size
|
| 350 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 351 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 352 |
+
return partitioned_numel, padding_numel
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 356 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 357 |
+
return
|
| 358 |
+
|
| 359 |
+
if debug:
|
| 360 |
+
for i in range(world_size):
|
| 361 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 362 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 363 |
+
|
| 364 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 365 |
+
wanted_params = len(frozen_param_shapes)
|
| 366 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 367 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 368 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 369 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 370 |
+
|
| 371 |
+
total_params = 0
|
| 372 |
+
total_numel = 0
|
| 373 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 374 |
+
total_params += 1
|
| 375 |
+
unpartitioned_numel = shape.numel()
|
| 376 |
+
total_numel += unpartitioned_numel
|
| 377 |
+
|
| 378 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 379 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 380 |
+
|
| 381 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 382 |
+
|
| 383 |
+
if debug:
|
| 384 |
+
print(
|
| 385 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 389 |
+
|
| 390 |
+
|
| 391 |
+
class GatheredTensor:
|
| 392 |
+
"""
|
| 393 |
+
A pseudo tensor that collects partitioned weights.
|
| 394 |
+
It is more memory efficient when there are multiple groups.
|
| 395 |
+
"""
|
| 396 |
+
|
| 397 |
+
def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
|
| 398 |
+
self.flat_groups = flat_groups
|
| 399 |
+
self.flat_groups_offset = flat_groups_offset
|
| 400 |
+
self.offset = offset
|
| 401 |
+
self.partitioned_numel = partitioned_numel
|
| 402 |
+
self.shape = shape
|
| 403 |
+
self.dtype = self.flat_groups[0][0].dtype
|
| 404 |
+
|
| 405 |
+
def contiguous(self):
|
| 406 |
+
"""
|
| 407 |
+
Merge partitioned weights from flat_groups into a single tensor.
|
| 408 |
+
"""
|
| 409 |
+
end_idx = self.offset + self.partitioned_numel
|
| 410 |
+
world_size = len(self.flat_groups)
|
| 411 |
+
pad_flat_param_chunks = []
|
| 412 |
+
|
| 413 |
+
for rank_i in range(world_size):
|
| 414 |
+
# for each rank, we need to collect weights from related group/groups
|
| 415 |
+
flat_groups_at_rank_i = self.flat_groups[rank_i]
|
| 416 |
+
start_group_id = None
|
| 417 |
+
end_group_id = None
|
| 418 |
+
for group_id in range(len(self.flat_groups_offset)):
|
| 419 |
+
if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
|
| 420 |
+
start_group_id = group_id
|
| 421 |
+
if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
|
| 422 |
+
end_group_id = group_id
|
| 423 |
+
break
|
| 424 |
+
# collect weights from related group/groups
|
| 425 |
+
for group_id in range(start_group_id, end_group_id + 1):
|
| 426 |
+
flat_tensor = flat_groups_at_rank_i[group_id]
|
| 427 |
+
start_offset = self.offset - self.flat_groups_offset[group_id]
|
| 428 |
+
end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
|
| 429 |
+
pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
|
| 430 |
+
|
| 431 |
+
# collect weights from all ranks
|
| 432 |
+
pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
|
| 433 |
+
param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
|
| 434 |
+
return param
|
| 435 |
+
|
| 436 |
+
|
| 437 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 438 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 439 |
+
avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
|
| 440 |
+
|
| 441 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 442 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 443 |
+
|
| 444 |
+
# merge list of dicts, preserving order
|
| 445 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 446 |
+
|
| 447 |
+
if debug:
|
| 448 |
+
for i in range(world_size):
|
| 449 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 450 |
+
|
| 451 |
+
wanted_params = len(param_shapes)
|
| 452 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 453 |
+
# not asserting if there is a mismatch due to possible padding
|
| 454 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 455 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 456 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 457 |
+
|
| 458 |
+
# params
|
| 459 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 460 |
+
# out-of-core computing solution
|
| 461 |
+
offset = 0
|
| 462 |
+
total_numel = 0
|
| 463 |
+
total_params = 0
|
| 464 |
+
flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
|
| 465 |
+
for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
|
| 466 |
+
unpartitioned_numel = shape.numel()
|
| 467 |
+
total_numel += unpartitioned_numel
|
| 468 |
+
total_params += 1
|
| 469 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 470 |
+
|
| 471 |
+
if debug:
|
| 472 |
+
print(
|
| 473 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 474 |
+
)
|
| 475 |
+
|
| 476 |
+
# memory efficient tensor
|
| 477 |
+
tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
|
| 478 |
+
state_dict[name] = tensor
|
| 479 |
+
offset += partitioned_numel
|
| 480 |
+
|
| 481 |
+
offset *= world_size
|
| 482 |
+
|
| 483 |
+
# Sanity check
|
| 484 |
+
if offset != avail_numel:
|
| 485 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 486 |
+
|
| 487 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 488 |
+
|
| 489 |
+
|
| 490 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 491 |
+
exclude_frozen_parameters):
|
| 492 |
+
state_dict = OrderedDict()
|
| 493 |
+
|
| 494 |
+
# buffers
|
| 495 |
+
buffers = zero_model_states[0].buffers
|
| 496 |
+
state_dict.update(buffers)
|
| 497 |
+
if debug:
|
| 498 |
+
print(f"added {len(buffers)} buffers")
|
| 499 |
+
|
| 500 |
+
if not exclude_frozen_parameters:
|
| 501 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 502 |
+
|
| 503 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 504 |
+
|
| 505 |
+
# recover shared parameters
|
| 506 |
+
for pair in zero_model_states[0].shared_params:
|
| 507 |
+
if pair[1] in state_dict:
|
| 508 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 509 |
+
|
| 510 |
+
return state_dict
|
| 511 |
+
|
| 512 |
+
|
| 513 |
+
def to_torch_tensor(state_dict, return_empty_tensor=False):
|
| 514 |
+
"""
|
| 515 |
+
Convert state_dict of GatheredTensor to torch tensor
|
| 516 |
+
"""
|
| 517 |
+
torch_state_dict = {}
|
| 518 |
+
converted_tensors = {}
|
| 519 |
+
for name, tensor in state_dict.items():
|
| 520 |
+
tensor_id = id(tensor)
|
| 521 |
+
if tensor_id in converted_tensors: # shared tensors
|
| 522 |
+
shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
|
| 523 |
+
torch_state_dict[name] = shared_tensor
|
| 524 |
+
else:
|
| 525 |
+
converted_tensors[tensor_id] = name
|
| 526 |
+
if return_empty_tensor:
|
| 527 |
+
torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
|
| 528 |
+
else:
|
| 529 |
+
torch_state_dict[name] = tensor.contiguous()
|
| 530 |
+
return torch_state_dict
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 534 |
+
tag=None,
|
| 535 |
+
exclude_frozen_parameters=False,
|
| 536 |
+
lazy_mode=False):
|
| 537 |
+
"""
|
| 538 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 539 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 540 |
+
via a model hub.
|
| 541 |
+
|
| 542 |
+
Args:
|
| 543 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 544 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
| 545 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 546 |
+
- ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
|
| 547 |
+
Convert the pesduo tensor to torch tensor by ``.contiguous()``
|
| 548 |
+
|
| 549 |
+
Returns:
|
| 550 |
+
- pytorch ``state_dict``
|
| 551 |
+
|
| 552 |
+
A typical usage might be ::
|
| 553 |
+
|
| 554 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 555 |
+
# do the training and checkpoint saving
|
| 556 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 557 |
+
model = model.cpu() # move to cpu
|
| 558 |
+
model.load_state_dict(state_dict)
|
| 559 |
+
# submit to model hub or save the model to share with others
|
| 560 |
+
|
| 561 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 562 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 563 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 564 |
+
|
| 565 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 566 |
+
|
| 567 |
+
Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
|
| 568 |
+
You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 569 |
+
the checkpoint. Or you can load state_dict in lazy mode ::
|
| 570 |
+
|
| 571 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 572 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
|
| 573 |
+
for name, lazy_tensor in state_dict.item():
|
| 574 |
+
tensor = lazy_tensor.contiguous() # to cpu
|
| 575 |
+
print(name, tensor)
|
| 576 |
+
# del tensor to release memory if it no longer in use
|
| 577 |
+
"""
|
| 578 |
+
if tag is None:
|
| 579 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 580 |
+
if os.path.isfile(latest_path):
|
| 581 |
+
with open(latest_path, 'r') as fd:
|
| 582 |
+
tag = fd.read().strip()
|
| 583 |
+
else:
|
| 584 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 585 |
+
|
| 586 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 587 |
+
|
| 588 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 589 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 590 |
+
|
| 591 |
+
state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
|
| 592 |
+
if lazy_mode:
|
| 593 |
+
return state_dict
|
| 594 |
+
else:
|
| 595 |
+
return to_torch_tensor(state_dict)
|
| 596 |
+
|
| 597 |
+
|
| 598 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
|
| 599 |
+
output_dir,
|
| 600 |
+
max_shard_size="5GB",
|
| 601 |
+
safe_serialization=False,
|
| 602 |
+
tag=None,
|
| 603 |
+
exclude_frozen_parameters=False):
|
| 604 |
+
"""
|
| 605 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 606 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 607 |
+
|
| 608 |
+
Args:
|
| 609 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 610 |
+
- ``output_dir``: directory to the pytorch fp32 state_dict output files
|
| 611 |
+
- ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
|
| 612 |
+
- ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
|
| 613 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 614 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 615 |
+
"""
|
| 616 |
+
|
| 617 |
+
# Dependency pre-check
|
| 618 |
+
if safe_serialization:
|
| 619 |
+
try:
|
| 620 |
+
from safetensors.torch import save_file
|
| 621 |
+
except ImportError:
|
| 622 |
+
print('If you want to use `safe_serialization`, please `pip install safetensors`')
|
| 623 |
+
raise
|
| 624 |
+
if max_shard_size is not None:
|
| 625 |
+
try:
|
| 626 |
+
from huggingface_hub import split_torch_state_dict_into_shards
|
| 627 |
+
except ImportError:
|
| 628 |
+
print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
|
| 629 |
+
raise
|
| 630 |
+
|
| 631 |
+
# Convert zero checkpoint to state_dict
|
| 632 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 633 |
+
tag,
|
| 634 |
+
exclude_frozen_parameters,
|
| 635 |
+
lazy_mode=True)
|
| 636 |
+
|
| 637 |
+
# Shard the model if it is too big.
|
| 638 |
+
weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
|
| 639 |
+
if max_shard_size is not None:
|
| 640 |
+
filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
|
| 641 |
+
# an memory-efficient approach for sharding
|
| 642 |
+
empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
|
| 643 |
+
state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
|
| 644 |
+
filename_pattern=filename_pattern,
|
| 645 |
+
max_shard_size=max_shard_size)
|
| 646 |
+
else:
|
| 647 |
+
from collections import namedtuple
|
| 648 |
+
StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
|
| 649 |
+
state_dict_split = StateDictSplit(is_sharded=False,
|
| 650 |
+
filename_to_tensors={weights_name: list(state_dict.keys())})
|
| 651 |
+
|
| 652 |
+
# Save the model by shard
|
| 653 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 654 |
+
filename_to_tensors = state_dict_split.filename_to_tensors.items()
|
| 655 |
+
for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
|
| 656 |
+
shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
|
| 657 |
+
shard_state_dict = to_torch_tensor(shard_state_dict)
|
| 658 |
+
output_path = os.path.join(output_dir, shard_file)
|
| 659 |
+
if safe_serialization:
|
| 660 |
+
save_file(shard_state_dict, output_path, metadata={"format": "pt"})
|
| 661 |
+
else:
|
| 662 |
+
torch.save(shard_state_dict, output_path)
|
| 663 |
+
# release the memory of current shard
|
| 664 |
+
for tensor_name in list(shard_state_dict.keys()):
|
| 665 |
+
del state_dict[tensor_name]
|
| 666 |
+
del shard_state_dict[tensor_name]
|
| 667 |
+
del shard_state_dict
|
| 668 |
+
gc.collect()
|
| 669 |
+
|
| 670 |
+
# Save index if sharded
|
| 671 |
+
if state_dict_split.is_sharded:
|
| 672 |
+
index = {
|
| 673 |
+
"metadata": state_dict_split.metadata,
|
| 674 |
+
"weight_map": state_dict_split.tensor_to_filename,
|
| 675 |
+
}
|
| 676 |
+
save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
|
| 677 |
+
save_index_file = os.path.join(output_dir, save_index_file)
|
| 678 |
+
with open(save_index_file, "w", encoding="utf-8") as f:
|
| 679 |
+
content = json.dumps(index, indent=2, sort_keys=True) + "\n"
|
| 680 |
+
f.write(content)
|
| 681 |
+
|
| 682 |
+
|
| 683 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 684 |
+
"""
|
| 685 |
+
1. Put the provided model to cpu
|
| 686 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 687 |
+
3. Load it into the provided model
|
| 688 |
+
|
| 689 |
+
Args:
|
| 690 |
+
- ``model``: the model object to update
|
| 691 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 692 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 693 |
+
|
| 694 |
+
Returns:
|
| 695 |
+
- ``model`: modified model
|
| 696 |
+
|
| 697 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 698 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 699 |
+
conveniently placed for you in the checkpoint folder.
|
| 700 |
+
|
| 701 |
+
A typical usage might be ::
|
| 702 |
+
|
| 703 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 704 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 705 |
+
# submit to model hub or save the model to share with others
|
| 706 |
+
|
| 707 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 708 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 709 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 710 |
+
|
| 711 |
+
"""
|
| 712 |
+
logger.info("Extracting fp32 weights")
|
| 713 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 714 |
+
|
| 715 |
+
logger.info("Overwriting model with fp32 weights")
|
| 716 |
+
model = model.cpu()
|
| 717 |
+
model.load_state_dict(state_dict, strict=False)
|
| 718 |
+
|
| 719 |
+
return model
|
| 720 |
+
|
| 721 |
+
|
| 722 |
+
if __name__ == "__main__":
|
| 723 |
+
parser = argparse.ArgumentParser()
|
| 724 |
+
parser.add_argument("checkpoint_dir",
|
| 725 |
+
type=str,
|
| 726 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 727 |
+
parser.add_argument("output_dir",
|
| 728 |
+
type=str,
|
| 729 |
+
help="directory to the pytorch fp32 state_dict output files"
|
| 730 |
+
"(e.g. path/checkpoint-12-output/)")
|
| 731 |
+
parser.add_argument(
|
| 732 |
+
"--max_shard_size",
|
| 733 |
+
type=str,
|
| 734 |
+
default="5GB",
|
| 735 |
+
help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
|
| 736 |
+
"lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
|
| 737 |
+
"We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
|
| 738 |
+
"without CPU OOM issues.")
|
| 739 |
+
parser.add_argument(
|
| 740 |
+
"--safe_serialization",
|
| 741 |
+
default=False,
|
| 742 |
+
action='store_true',
|
| 743 |
+
help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
|
| 744 |
+
parser.add_argument("-t",
|
| 745 |
+
"--tag",
|
| 746 |
+
type=str,
|
| 747 |
+
default=None,
|
| 748 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 749 |
+
parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
|
| 750 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 751 |
+
args = parser.parse_args()
|
| 752 |
+
|
| 753 |
+
debug = args.debug
|
| 754 |
+
|
| 755 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
|
| 756 |
+
args.output_dir,
|
| 757 |
+
max_shard_size=args.max_shard_size,
|
| 758 |
+
safe_serialization=args.safe_serialization,
|
| 759 |
+
tag=args.tag,
|
| 760 |
+
exclude_frozen_parameters=args.exclude_frozen_parameters)
|