Upload continuous evaluation checkpoints, batch 1
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +3 -0
- continuous/qwen25/eupe_vit_b/pretrain/config.json +87 -0
- continuous/qwen25/eupe_vit_b/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/eupe_vit_b/pretrain/trainer_state.json +421 -0
- continuous/qwen25/eupe_vit_s/finetune/added_tokens.json +24 -0
- continuous/qwen25/eupe_vit_s/finetune/chat_template.jinja +54 -0
- continuous/qwen25/eupe_vit_s/finetune/config.json +87 -0
- continuous/qwen25/eupe_vit_s/finetune/generation_config.json +13 -0
- continuous/qwen25/eupe_vit_s/finetune/merges.txt +0 -0
- continuous/qwen25/eupe_vit_s/finetune/model.safetensors +3 -0
- continuous/qwen25/eupe_vit_s/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/eupe_vit_s/finetune/tokenizer.json +3 -0
- continuous/qwen25/eupe_vit_s/finetune/tokenizer_config.json +208 -0
- continuous/qwen25/eupe_vit_s/finetune/trainer_state.json +1856 -0
- continuous/qwen25/eupe_vit_s/finetune/training_args.bin +3 -0
- continuous/qwen25/eupe_vit_s/finetune/vocab.json +0 -0
- continuous/qwen25/eupe_vit_s/pretrain/config.json +87 -0
- continuous/qwen25/eupe_vit_s/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/eupe_vit_s/pretrain/trainer_state.json +421 -0
- continuous/qwen25/eupe_vit_t/finetune/added_tokens.json +24 -0
- continuous/qwen25/eupe_vit_t/finetune/chat_template.jinja +54 -0
- continuous/qwen25/eupe_vit_t/finetune/config.json +87 -0
- continuous/qwen25/eupe_vit_t/finetune/generation_config.json +13 -0
- continuous/qwen25/eupe_vit_t/finetune/merges.txt +0 -0
- continuous/qwen25/eupe_vit_t/finetune/model.safetensors +3 -0
- continuous/qwen25/eupe_vit_t/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/eupe_vit_t/finetune/tokenizer.json +3 -0
- continuous/qwen25/eupe_vit_t/finetune/tokenizer_config.json +208 -0
- continuous/qwen25/eupe_vit_t/finetune/trainer_state.json +1856 -0
- continuous/qwen25/eupe_vit_t/finetune/training_args.bin +3 -0
- continuous/qwen25/eupe_vit_t/finetune/vocab.json +0 -0
- continuous/qwen25/eupe_vit_t/pretrain/config.json +87 -0
- continuous/qwen25/eupe_vit_t/pretrain/trainer_state.json +421 -0
- continuous/qwen25/ijepa_vith14/finetune/added_tokens.json +24 -0
- continuous/qwen25/ijepa_vith14/finetune/chat_template.jinja +54 -0
- continuous/qwen25/ijepa_vith14/finetune/config.json +87 -0
- continuous/qwen25/ijepa_vith14/finetune/generation_config.json +13 -0
- continuous/qwen25/ijepa_vith14/finetune/merges.txt +0 -0
- continuous/qwen25/ijepa_vith14/finetune/model.safetensors +3 -0
- continuous/qwen25/ijepa_vith14/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/ijepa_vith14/finetune/tokenizer.json +3 -0
- continuous/qwen25/ijepa_vith14/finetune/tokenizer_config.json +208 -0
- continuous/qwen25/ijepa_vith14/finetune/trainer_state.json +1856 -0
- continuous/qwen25/ijepa_vith14/finetune/training_args.bin +3 -0
- continuous/qwen25/ijepa_vith14/finetune/vocab.json +0 -0
- continuous/qwen25/ijepa_vith14/pretrain/config.json +87 -0
- continuous/qwen25/ijepa_vith14/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/ijepa_vith14/pretrain/trainer_state.json +421 -0
- continuous/qwen25/mc1_b16_224_2.5b/finetune/added_tokens.json +24 -0
- continuous/qwen25/mc1_b16_224_2.5b/finetune/config.json +57 -0
.gitattributes
CHANGED
|
@@ -48,3 +48,6 @@ continuous/qwen25/eupe_convnext_b/finetune/tokenizer.json filter=lfs diff=lfs me
|
|
| 48 |
continuous/qwen25/eupe_convnext_s/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 49 |
continuous/qwen25/eupe_convnext_t/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 50 |
continuous/qwen25/eupe_vit_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
continuous/qwen25/eupe_convnext_s/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 49 |
continuous/qwen25/eupe_convnext_t/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 50 |
continuous/qwen25/eupe_vit_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
continuous/qwen25/eupe_vit_s/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
continuous/qwen25/eupe_vit_t/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
continuous/qwen25/ijepa_vith14/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
continuous/qwen25/eupe_vit_b/pretrain/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen2ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 768,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": null,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-B/EUPE-ViT-B.pt",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/eupe_vit_b/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b39900c991f5fc07af7f5c025004ff47ab15da5a8af1ce0980273c798ee350bd
|
| 3 |
+
size 14170168
|
continuous/qwen25/eupe_vit_b/pretrain/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 545,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01834862385321101,
|
| 14 |
+
"grad_norm": 1.1379262208938599,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.8786,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.8404291868209839,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.9769,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.2107398509979248,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.4903,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.18311628699302673,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.3043,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.24167323112487793,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.2604,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.17213773727416992,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.2194,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.2814987897872925,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.183,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.11983971297740936,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 3.1574,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.30033284425735474,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 3.1303,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.14500291645526886,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 3.1141,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.16436219215393066,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 3.0896,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.17327828705310822,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 3.0777,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.1351078599691391,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 3.0461,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.16872438788414001,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 3.0325,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.179401695728302,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 3.0261,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.1830599159002304,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 3.0066,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.17815729975700378,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.9988,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.20790863037109375,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.9784,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.17421481013298035,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.9734,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.13976486027240753,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.9474,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.1535816341638565,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.9389,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.1641007512807846,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.9443,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.1455346643924713,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.9392,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.19196979701519012,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.9241,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.1843026876449585,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.9096,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.13931746780872345,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.9207,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.14458073675632477,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.8998,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.1333315074443817,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.9003,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.146688312292099,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.8782,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.13721095025539398,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.877,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.13452450931072235,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.8861,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.13342800736427307,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.8795,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.1756356656551361,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.8604,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.14014668762683868,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.865,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.12824876606464386,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.8663,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.16090673208236694,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.8673,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.13102516531944275,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.8569,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.1255619376897812,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.8661,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.16729342937469482,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.8559,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.13838130235671997,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.8478,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.12839056551456451,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.8503,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.13885152339935303,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.843,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.12069682776927948,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.8437,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.12974156439304352,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.8499,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.1198401153087616,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.8495,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.12066667526960373,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.8359,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.11729896068572998,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.8279,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.1309446096420288,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.836,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.12500585615634918,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.8402,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.10695896297693253,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.8483,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.11506281048059464,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.8371,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.10993779450654984,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.8309,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.11368147283792496,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.832,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.11225897818803787,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.836,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.783114240595067e+17,
|
| 394 |
+
"train_loss": 3.0069172412977307,
|
| 395 |
+
"train_runtime": 5269.6944,
|
| 396 |
+
"train_samples_per_second": 105.913,
|
| 397 |
+
"train_steps_per_second": 0.103
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 33142912,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 2.783114240595067e+17,
|
| 418 |
+
"train_batch_size": 64,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# 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>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\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" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
continuous/qwen25/eupe_vit_s/finetune/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwenForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 384,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": 0.0001,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-S/EUPE-ViT-S.pt",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"repetition_penalty": 1.1,
|
| 9 |
+
"temperature": 0.7,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.8,
|
| 12 |
+
"transformers_version": "4.57.6"
|
| 13 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/eupe_vit_s/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e84c9f27f8a760d3ac3d72265cb27d50efe71505dad422cb96461826d7b1970
|
| 3 |
+
size 3609277952
|
continuous/qwen25/eupe_vit_s/finetune/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 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
continuous/qwen25/eupe_vit_s/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,208 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": null,
|
| 198 |
+
"clean_up_tokenization_spaces": false,
|
| 199 |
+
"eos_token": "<|im_end|>",
|
| 200 |
+
"errors": "replace",
|
| 201 |
+
"extra_special_tokens": {},
|
| 202 |
+
"model_max_length": 8192,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"padding_side": "right",
|
| 205 |
+
"split_special_tokens": false,
|
| 206 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 207 |
+
"unk_token": null
|
| 208 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/trainer_state.json
ADDED
|
@@ -0,0 +1,1856 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9999037443449803,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2597,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.003850226200789296,
|
| 14 |
+
"grad_norm": 15.263704299926758,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.9313,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 14.893720626831055,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.6807,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 3.9490556716918945,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.183,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.9317399859428406,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9634,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.36062106490135193,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.9251,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.3021993637084961,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.8804,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.23236793279647827,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.855,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.24412305653095245,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.864,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.21918994188308716,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.8456,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.2282455563545227,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.827,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.20932689309120178,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8161,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.23376420140266418,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8286,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.21189942955970764,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8205,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.2377437800168991,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8209,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.24948836863040924,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.8036,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.22915641963481903,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8106,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.2000190019607544,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8118,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.2333904653787613,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8391,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.22952428460121155,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8295,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.25150394439697266,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8242,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.21619777381420135,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8125,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.22336266934871674,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8028,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.22455263137817383,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.8047,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.23729684948921204,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.7993,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.23010069131851196,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8137,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.2193240076303482,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8137,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.21697057783603668,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.8017,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.20915305614471436,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8019,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.2219613790512085,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8217,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.2107265591621399,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.7821,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.21815651655197144,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.7924,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.2233743816614151,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.7965,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.2264641970396042,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.7942,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.22050058841705322,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.8077,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.2301616370677948,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.806,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.21736596524715424,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.7929,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.2200148105621338,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8179,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.2208910882472992,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.7833,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.22444000840187073,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.7882,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.20860826969146729,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.7945,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.2132520228624344,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.7864,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.2257131040096283,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.7983,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.22059805691242218,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.7896,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.20925140380859375,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.7894,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.2211647480726242,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7769,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.2215186208486557,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.7822,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.21106676757335663,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.7904,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.22351977229118347,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.7899,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.20051996409893036,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7756,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.20469941198825836,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.7845,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.22042877972126007,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7757,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.2343902885913849,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.805,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.22617167234420776,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.7811,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.2363426834344864,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.7916,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.2220606505870819,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.7934,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.20808298885822296,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.7981,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.2315520942211151,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8094,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.22751688957214355,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.786,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.21750390529632568,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.7981,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.2220136970281601,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.7895,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.23600082099437714,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.8094,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.21708029508590698,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.7932,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.2288910299539566,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8093,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.21018525958061218,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.7986,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.2010924369096756,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7572,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.2374844253063202,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.7931,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.23863796889781952,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.7979,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.20692485570907593,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7689,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.2127869576215744,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.792,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.22369055449962616,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.7959,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.21494464576244354,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7775,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.2042323499917984,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.7824,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.21424296498298645,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.7865,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.23784619569778442,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.778,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.2065483033657074,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.7754,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.2201537936925888,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7686,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.21044692397117615,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.786,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.20547325909137726,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.7817,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.2222902774810791,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.7695,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.21216323971748352,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.7959,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.22924096882343292,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.7886,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.211440309882164,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.7833,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.21001508831977844,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7682,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.2326323688030243,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7733,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.2128707766532898,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.7833,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.21640120446681976,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7698,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.20984600484371185,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.7791,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.20227999985218048,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7736,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.20488876104354858,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.7728,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.19232019782066345,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.7788,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.2220333367586136,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.7646,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.23245765268802643,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7652,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.2344382107257843,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7542,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.24010811746120453,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.787,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.22194913029670715,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8005,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.22699494659900665,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.7865,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.2195494920015335,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7672,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.21574075520038605,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7721,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.2245539426803589,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7582,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.21354027092456818,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.7714,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.21170304715633392,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.7967,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.23205776512622833,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.789,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.21180665493011475,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.7866,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.22086332738399506,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.777,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.19933195412158966,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.7735,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.20081354677677155,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.7804,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.21739742159843445,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7699,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.21121461689472198,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.7947,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.23505374789237976,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.7769,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.21818296611309052,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.7912,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.22564099729061127,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.7875,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.23938733339309692,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.7836,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.2186473309993744,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7615,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.2140062004327774,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.7819,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.21836358308792114,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.775,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.2539863586425781,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.7969,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.2157716006040573,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8093,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.2209499031305313,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.7809,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.22332672774791718,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7631,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.22496244311332703,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7731,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.2241455316543579,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7761,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.22217898070812225,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.7921,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.21244345605373383,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.7965,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.20747177302837372,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7646,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.20569349825382233,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7639,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2365402728319168,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7719,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.24063795804977417,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7566,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.21598301827907562,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7643,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.2204660177230835,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7749,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.21137036383152008,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.7763,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.2114977240562439,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.7875,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.22539778053760529,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7601,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.24463261663913727,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7692,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2233595997095108,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7737,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.21795928478240967,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7659,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.24439865350723267,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.7857,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.22338375449180603,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.7843,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.2086922824382782,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.7902,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.22879794239997864,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7701,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.21302804350852966,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7696,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.2265365719795227,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.7827,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.23044927418231964,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7697,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.23218561708927155,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.7846,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.22382375597953796,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.7877,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.23098362982273102,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.77,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.22549070417881012,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7748,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.21780474483966827,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7715,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.22511017322540283,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7673,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.22080263495445251,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7734,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.20530852675437927,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.7819,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.2222430408000946,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.7762,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.21297000348567963,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.7847,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.21803684532642365,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.7871,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.20803669095039368,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.79,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.21654678881168365,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.7777,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.21986615657806396,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.7941,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.222995787858963,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.7884,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.2351967841386795,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.779,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.19251620769500732,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.7794,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.20997776091098785,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7679,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.20112276077270508,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7498,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.2175419181585312,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7502,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.215488001704216,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.7817,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.19199304282665253,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7701,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.21977180242538452,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.7798,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.22105354070663452,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7603,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.22764694690704346,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.7791,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.2303556501865387,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.7776,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.22310180962085724,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7661,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.1955786645412445,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7675,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.22327177226543427,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.7785,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.22454671561717987,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7686,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.22163774073123932,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7749,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.21835625171661377,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7742,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.20705139636993408,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7721,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.2077990621328354,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.761,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.2254655659198761,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8032,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.22568686306476593,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.7887,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.2176145315170288,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.7993,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.21538953483104706,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7623,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.22960145771503448,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7556,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.205283060669899,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.7776,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.2119828462600708,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7665,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.2272314727306366,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7734,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.23265641927719116,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.783,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.21637938916683197,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.7783,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.2004072070121765,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7741,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.2110900729894638,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.7813,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.22476541996002197,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7643,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.22257445752620697,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.7742,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.209440678358078,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.7731,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.21846361458301544,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7655,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.22390176355838776,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.7836,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.2296438217163086,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.7893,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.22262440621852875,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.7747,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.23362943530082703,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.7946,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.2138333022594452,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.7815,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.22662581503391266,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.784,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.22641238570213318,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.7739,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.21112361550331116,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7666,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.21445414423942566,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7762,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.20640626549720764,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7622,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.2161131352186203,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7505,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.2156892865896225,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7658,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.20677855610847473,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7691,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.1997710019350052,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7558,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.23452386260032654,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.7781,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.2068513184785843,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7593,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.21419821679592133,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.7812,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.2199387401342392,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7691,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.233067125082016,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.7773,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.21816398203372955,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.7791,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.23576945066452026,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7649,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.2259545922279358,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.7701,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.23111635446548462,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7624,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.21537324786186218,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.7795,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.21445803344249725,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.782,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.21095149219036102,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7708,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.21301423013210297,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.7807,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.23711547255516052,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7735,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.21992351114749908,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7441,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.20252276957035065,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.7776,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.2229768931865692,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7726,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.2058701366186142,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.7826,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.20965664088726044,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.7682,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.21890458464622498,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7583,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.2186364084482193,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7665,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.21771538257598877,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7738,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.2177412062883377,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.7811,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.20117653906345367,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7591,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.19972853362560272,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7731,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.2107749879360199,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.777,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.2118757963180542,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.7804,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.21396999061107635,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.7756,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.2137463241815567,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7576,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.21115340292453766,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7702,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.23302105069160461,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7627,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.2162923663854599,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.773,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.21746347844600677,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.7809,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.2333754152059555,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.7816,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.20973685383796692,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7612,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.21720494329929352,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.7876,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.22656171023845673,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.7808,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.2252018302679062,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.7716,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.23262758553028107,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.77,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.21629488468170166,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.7873,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.23171871900558472,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.7844,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.22099395096302032,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7652,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.2154410481452942,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.7728,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.214185893535614,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7676,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.2161446064710617,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7593,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.21495893597602844,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.7793,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.22100506722927094,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7665,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.2108873873949051,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7677,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.2122345119714737,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.7777,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.20869189500808716,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7713,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.21524721384048462,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7605,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.2102876901626587,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7529,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.22539763152599335,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.7834,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.3547648496440443e+18,
|
| 1829 |
+
"train_loss": 0.7928232331435682,
|
| 1830 |
+
"train_runtime": 10091.8781,
|
| 1831 |
+
"train_samples_per_second": 65.885,
|
| 1832 |
+
"train_steps_per_second": 0.257
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 418855352,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 3.3547648496440443e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/eupe_vit_s/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62d92d899b899fbcdc014ac2ae0be26c99a2939b9223598b6b4660f9b7d9a03a
|
| 3 |
+
size 6264
|
continuous/qwen25/eupe_vit_s/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/eupe_vit_s/pretrain/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen2ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 384,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": null,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-S/EUPE-ViT-S.pt",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/eupe_vit_s/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60497d2801dfae3dcb5ef4bc423a29b8c7632006f91734e9ae3312d622199afb
|
| 3 |
+
size 11810872
|
continuous/qwen25/eupe_vit_s/pretrain/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 545,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01834862385321101,
|
| 14 |
+
"grad_norm": 5.612421989440918,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.8473,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.349037230014801,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.747,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.12441191077232361,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.4029,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.10554260015487671,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.2507,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.1428859829902649,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.1815,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.09803934395313263,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.1303,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.1782945841550827,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.0909,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.10349910706281662,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 3.0623,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.11385659128427505,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 3.033,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.11282248049974442,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 3.019,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.16748358309268951,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 2.9942,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.1715393364429474,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 2.9887,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.11907421052455902,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 2.9605,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.17318814992904663,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 2.9545,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.1094144880771637,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 2.9536,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.10171601176261902,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 2.9387,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.1319226622581482,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.9337,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.12991686165332794,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.9193,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.1075221449136734,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.9188,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.1486317664384842,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.9018,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.11178604513406754,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.8999,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.11745817959308624,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.9092,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.18738628923892975,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.906,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.10819970816373825,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.8952,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.11976604163646698,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.8834,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.13871832191944122,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.8961,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.12912052869796753,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.8751,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.110613614320755,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.8814,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.11309967935085297,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.8619,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.10402233153581619,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.8651,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.09551090002059937,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.8733,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.10413948446512222,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.869,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.10761623829603195,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.8519,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.12687504291534424,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.8567,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.09913207590579987,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.8611,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.12111896276473999,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.8621,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.09930725395679474,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.8527,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.08716098964214325,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.8639,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.1251896470785141,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.851,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.0913405492901802,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.8472,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.08800724893808365,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.8455,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.08967752754688263,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.8399,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.08932823687791824,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.8442,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.10521621257066727,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.852,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.09559813141822815,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.849,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.09309468418359756,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.8359,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.09236087650060654,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.8307,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.09678466618061066,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.8362,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.09556855261325836,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.8429,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.08732381463050842,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.8498,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.08921365439891815,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.8393,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.09148542582988739,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.8336,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.09691683948040009,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.8358,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.08641217648983002,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.835,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.6545363683691725e+17,
|
| 394 |
+
"train_loss": 2.970720763600201,
|
| 395 |
+
"train_runtime": 5260.2168,
|
| 396 |
+
"train_samples_per_second": 106.104,
|
| 397 |
+
"train_steps_per_second": 0.104
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 33142912,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 2.6545363683691725e+17,
|
| 418 |
+
"train_batch_size": 64,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# 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>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\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" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
continuous/qwen25/eupe_vit_t/finetune/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwenForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 192,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": 0.0001,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-T/EUPE-ViT-T.pt",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"repetition_penalty": 1.1,
|
| 9 |
+
"temperature": 0.7,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.8,
|
| 12 |
+
"transformers_version": "4.57.6"
|
| 13 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/eupe_vit_t/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d70015155c43feeecc76bc1853559eafcf469d000f6867b4ab5c3822c11abfe8
|
| 3 |
+
size 3575866416
|
continuous/qwen25/eupe_vit_t/finetune/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 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
continuous/qwen25/eupe_vit_t/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,208 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": null,
|
| 198 |
+
"clean_up_tokenization_spaces": false,
|
| 199 |
+
"eos_token": "<|im_end|>",
|
| 200 |
+
"errors": "replace",
|
| 201 |
+
"extra_special_tokens": {},
|
| 202 |
+
"model_max_length": 8192,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"padding_side": "right",
|
| 205 |
+
"split_special_tokens": false,
|
| 206 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 207 |
+
"unk_token": null
|
| 208 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/trainer_state.json
ADDED
|
@@ -0,0 +1,1856 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9999037443449803,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2597,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.003850226200789296,
|
| 14 |
+
"grad_norm": 13.461159706115723,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.8659,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 11.289884567260742,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.5972,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 3.3776795864105225,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.1531,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.7393693327903748,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9801,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.3550540506839752,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.9475,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.2969323992729187,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.9026,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.24133804440498352,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.8751,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.2458762526512146,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.8848,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.21807630360126495,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.865,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.22817905247211456,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8462,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.2121012657880783,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8352,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.23818176984786987,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8485,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.216318279504776,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8396,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.23711660504341125,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8401,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.24504733085632324,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.8228,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.2265852391719818,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8296,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.2034302055835724,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8312,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.2368393987417221,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8573,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.235277459025383,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8487,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.25363680720329285,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8431,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.21695558726787567,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.832,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.23445886373519897,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8227,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.22054043412208557,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.8253,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.22770392894744873,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.8181,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.22974155843257904,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8337,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.22814954817295074,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8333,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.21700260043144226,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.8213,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.20700694620609283,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8228,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.23301342129707336,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8422,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.20515602827072144,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.8009,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.2184620201587677,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.8121,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.22031068801879883,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.8165,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.2274760603904724,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.8156,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.2237875908613205,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.8265,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.22644175589084625,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.8264,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.22197556495666504,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.8114,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.21961289644241333,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8378,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.21255382895469666,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.8037,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.23193620145320892,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.8081,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.20273344218730927,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.8148,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.21061594784259796,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.8075,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.22430060803890228,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.8189,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.22110635042190552,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.8091,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.20654495060443878,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.8107,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.2201339602470398,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7975,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.22088362276554108,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.8029,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.21152372658252716,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.8125,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.22094152867794037,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.8112,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.20246224105358124,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7966,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.19715726375579834,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.8046,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.22382096946239471,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7973,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.2363816797733307,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.8268,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.22679375112056732,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.8013,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.22710898518562317,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.8119,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.2158307582139969,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.8146,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.2058659791946411,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.8192,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.22984099388122559,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8313,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.2344251275062561,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.8069,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.216031014919281,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.8199,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.21512436866760254,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.8114,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.2278410792350769,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.831,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.21194107830524445,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.8138,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.2268293797969818,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8322,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.21663610637187958,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.8192,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.21024824678897858,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7782,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.23823201656341553,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.8143,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.24769271910190582,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.8187,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.20600733160972595,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7906,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.21308773756027222,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.8148,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.22094488143920898,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.8183,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.21586431562900543,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7985,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.1992930918931961,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.804,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.21420906484127045,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.8083,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.2225823998451233,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.8002,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.20217283070087433,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.7971,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.21413494646549225,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7907,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.20422127842903137,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.8073,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.20040535926818848,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.804,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.2213214486837387,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.792,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.20986445248126984,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.8186,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.22112280130386353,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.8117,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.20809631049633026,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.8063,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.20797725021839142,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7915,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.22937579452991486,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7958,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.20757688581943512,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.8047,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.2155638486146927,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7931,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.2065477818250656,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.8015,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.20052799582481384,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7951,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.20458169281482697,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.795,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.19189774990081787,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.8026,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.22003473341464996,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.7865,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.221382737159729,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7878,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.23537157475948334,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7767,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.23904231190681458,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.809,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.22327707707881927,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8235,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.22470727562904358,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.8099,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.21491284668445587,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7915,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.20553696155548096,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7947,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.2206290364265442,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7816,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.21539047360420227,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.7946,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.20654897391796112,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.8182,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.2295886129140854,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.8136,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.20202083885669708,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.8103,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.21547256410121918,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.8016,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.1962355077266693,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.7984,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.19663064181804657,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.8037,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.21360380947589874,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7941,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.21080943942070007,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.8183,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.2352708876132965,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.8001,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.21371538937091827,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.8143,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2194889634847641,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.8125,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.23415403068065643,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.8074,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.2167832851409912,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7845,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.2082771360874176,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.8063,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.21120668947696686,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.7995,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.2443781942129135,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.8214,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.21239063143730164,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8344,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.2278509885072708,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.8043,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.21661382913589478,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7867,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.2203165888786316,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7966,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.22846537828445435,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7991,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.22484742105007172,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.8161,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.20509076118469238,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.8189,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.2001366913318634,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7879,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.20261621475219727,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7884,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.22927725315093994,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7957,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.2381802350282669,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7815,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.21262049674987793,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7879,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.2154548168182373,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7997,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.20302622020244598,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.8005,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.20600144565105438,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.8106,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.21956469118595123,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7839,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.23590917885303497,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.794,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2173629254102707,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7987,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.2091638594865799,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7902,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.24006107449531555,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.8095,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.21573616564273834,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.8085,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.19877302646636963,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.8136,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.2327021360397339,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7941,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.20471996068954468,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7935,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.22001561522483826,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.8077,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.2254607379436493,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7945,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.22590789198875427,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.8097,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.22284121811389923,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.8107,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.2246236503124237,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.7944,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.21901895105838776,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7986,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.21204160153865814,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7965,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.22383931279182434,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7909,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.2047976851463318,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7988,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.20047461986541748,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.8067,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.21688386797904968,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.8002,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.21117524802684784,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.8096,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.21311995387077332,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.8114,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.2071188986301422,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.8143,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.22191154956817627,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.8023,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.2152388095855713,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.8193,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.22587883472442627,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.8139,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.2307194173336029,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.8032,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.18939785659313202,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.8044,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.20617568492889404,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7938,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.1992744505405426,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7744,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.21673429012298584,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.774,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.21431250870227814,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.8063,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.18903523683547974,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7949,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.21716414391994476,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.8046,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.21132394671440125,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7848,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.22363072633743286,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.8044,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.22127237915992737,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.8019,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.21975743770599365,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7914,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.19061823189258575,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7931,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.21676869690418243,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.8031,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.2237246334552765,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7932,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.212800994515419,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7999,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.2169068604707718,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7995,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.2041679471731186,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7964,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.19894583523273468,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7872,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.22105860710144043,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8278,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.2218465805053711,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.8142,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.2073533535003662,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.8244,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.21041510999202728,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7869,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.22459764778614044,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7803,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.20919284224510193,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.8025,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.2044724076986313,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7913,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.2217891961336136,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7981,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.2257901430130005,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.8065,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.21271908283233643,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.8031,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.19536136090755463,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7977,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.20260578393936157,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.8068,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.2270425409078598,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7899,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.21941035985946655,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.8006,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.20361357927322388,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.7993,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.21181435883045197,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7918,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.2200392633676529,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.8107,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.22302794456481934,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.8136,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.2199358195066452,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.799,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.23151744902133942,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.82,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.21074360609054565,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.8062,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.23118513822555542,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.808,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.22614158689975739,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.798,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.21166346967220306,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.791,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.21542350947856903,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7999,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.20001953840255737,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7879,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.2100607454776764,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7747,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.2116590440273285,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7918,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.20311373472213745,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7953,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.1964741349220276,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7807,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.2295994907617569,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.8038,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.2170003354549408,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7849,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.20032645761966705,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.8063,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.2123996615409851,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7965,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.21669436991214752,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.8027,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.2095137983560562,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.8033,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.23304133117198944,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7907,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.21755537390708923,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.7943,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.2327924519777298,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7869,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.2106851488351822,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.8043,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.2110110968351364,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.8066,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.2080550342798233,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7948,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.20865361392498016,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.807,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.2289067804813385,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7984,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.21421346068382263,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7679,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.19590617716312408,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.8028,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.2114134579896927,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7972,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.20378540456295013,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.807,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.20594578981399536,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.7945,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.21080255508422852,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7825,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.2109915018081665,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7925,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.20986755192279816,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7989,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.21465882658958435,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.8077,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.1984810084104538,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7851,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.19682733714580536,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7976,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.20756617188453674,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.8017,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.2060251086950302,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.8052,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.20931079983711243,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.8014,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.2131694257259369,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7834,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.20518934726715088,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7949,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.2289833128452301,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7884,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.2140902727842331,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7978,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.2055724561214447,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.805,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.22757397592067719,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.8061,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.20862558484077454,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7864,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.21108709275722504,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.8131,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.223557710647583,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.8058,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.22114171087741852,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.7974,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.22645270824432373,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7959,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.2127862572669983,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.8136,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.22355836629867554,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.8088,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.21113921701908112,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7899,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.21291126310825348,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.7977,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.21146701276302338,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7934,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.2145322859287262,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7849,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.21674837172031403,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.8044,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.21389615535736084,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.792,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.2000604271888733,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7923,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.20141085982322693,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.8033,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.20697586238384247,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7968,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.2138867974281311,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7847,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.20393231511116028,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7778,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.21932105720043182,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.8073,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.313539680141574e+18,
|
| 1829 |
+
"train_loss": 0.8151919168832902,
|
| 1830 |
+
"train_runtime": 10086.3636,
|
| 1831 |
+
"train_samples_per_second": 65.921,
|
| 1832 |
+
"train_steps_per_second": 0.257
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 418855352,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 3.313539680141574e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/eupe_vit_t/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbd4d3cd451c15fcde72e4164279ffc2a658e7478ab4f04dbb91cedc1fd5a459
|
| 3 |
+
size 6264
|
continuous/qwen25/eupe_vit_t/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/eupe_vit_t/pretrain/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen2ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 192,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": null,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ViT-T/EUPE-ViT-T.pt",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/eupe_vit_t/pretrain/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 545,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01834862385321101,
|
| 14 |
+
"grad_norm": 1.015489101409912,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.8281,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.41837459802627563,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.9182,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.2522580921649933,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.4585,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.16748110949993134,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.3485,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.10977907478809357,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.3152,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.18595045804977417,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.2831,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.11838805675506592,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.2963,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.13766202330589294,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 3.2861,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.11473467946052551,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 3.2537,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.0951407253742218,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 3.2465,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.07488645613193512,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 3.2292,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.12827281653881073,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 3.2301,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.10331003367900848,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 3.2092,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.09305416792631149,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 3.2057,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.16418272256851196,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 3.208,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.1803768277168274,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 3.1971,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.10056817531585693,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 3.1969,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.07766424119472504,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 3.1851,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.09608001261949539,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 3.1878,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.08199429512023926,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 3.1688,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.11999441683292389,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 3.1694,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.10026830434799194,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 3.1779,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.15399561822414398,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 3.1795,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.08229731023311615,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 3.1681,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.10165657848119736,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 3.1598,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.08827485144138336,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 3.1724,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.09277576953172684,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 3.1575,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.09415962547063828,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 3.1605,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.10209429264068604,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 3.1447,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.09190699458122253,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 3.1458,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.09163729101419449,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 3.1557,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.08985079079866409,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 3.1544,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.12336169183254242,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 3.1405,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.10266783833503723,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 3.1441,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.09781866520643234,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 3.1536,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.10414116084575653,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 3.1507,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.09013304114341736,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 3.1392,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.11584986001253128,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 3.153,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.11197736114263535,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 3.1457,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.08574221283197403,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 3.1389,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.08761900663375854,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 3.1407,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.10369107127189636,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 3.1326,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.08590344339609146,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 3.1381,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.09561969339847565,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 3.1487,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.08258525282144547,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 3.145,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.07827068120241165,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 3.1356,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.09977085888385773,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 3.1253,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.08129765838384628,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 3.134,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.0745750367641449,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 3.1398,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.06724604964256287,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 3.1465,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.09364260733127594,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 3.1333,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.07172564417123795,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 3.1292,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.07236991077661514,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 3.1328,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.0725344866514206,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 3.1359,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.6219159951192883e+17,
|
| 394 |
+
"train_loss": 3.2269366238095345,
|
| 395 |
+
"train_runtime": 5304.5376,
|
| 396 |
+
"train_samples_per_second": 105.217,
|
| 397 |
+
"train_steps_per_second": 0.103
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 33142912,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 2.6219159951192883e+17,
|
| 418 |
+
"train_batch_size": 64,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# 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>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\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" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
continuous/qwen25/ijepa_vith14/finetune/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwenForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 1280,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": 0.0001,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:ijepa",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/ijepa/ijepa_vith.pth",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"repetition_penalty": 1.1,
|
| 9 |
+
"temperature": 0.7,
|
| 10 |
+
"top_k": 20,
|
| 11 |
+
"top_p": 0.8,
|
| 12 |
+
"transformers_version": "4.57.6"
|
| 13 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/ijepa_vith14/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26b36be88407910a601c55384928613328e253e529f844e9dc9a1bfa39d459b0
|
| 3 |
+
size 4833104016
|
continuous/qwen25/ijepa_vith14/finetune/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 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
|
| 3 |
+
size 11421896
|
continuous/qwen25/ijepa_vith14/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,208 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
},
|
| 182 |
+
"additional_special_tokens": [
|
| 183 |
+
"<|im_start|>",
|
| 184 |
+
"<|im_end|>",
|
| 185 |
+
"<|object_ref_start|>",
|
| 186 |
+
"<|object_ref_end|>",
|
| 187 |
+
"<|box_start|>",
|
| 188 |
+
"<|box_end|>",
|
| 189 |
+
"<|quad_start|>",
|
| 190 |
+
"<|quad_end|>",
|
| 191 |
+
"<|vision_start|>",
|
| 192 |
+
"<|vision_end|>",
|
| 193 |
+
"<|vision_pad|>",
|
| 194 |
+
"<|image_pad|>",
|
| 195 |
+
"<|video_pad|>"
|
| 196 |
+
],
|
| 197 |
+
"bos_token": null,
|
| 198 |
+
"clean_up_tokenization_spaces": false,
|
| 199 |
+
"eos_token": "<|im_end|>",
|
| 200 |
+
"errors": "replace",
|
| 201 |
+
"extra_special_tokens": {},
|
| 202 |
+
"model_max_length": 8192,
|
| 203 |
+
"pad_token": "<|endoftext|>",
|
| 204 |
+
"padding_side": "right",
|
| 205 |
+
"split_special_tokens": false,
|
| 206 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 207 |
+
"unk_token": null
|
| 208 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/trainer_state.json
ADDED
|
@@ -0,0 +1,1856 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9999037443449803,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2597,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.003850226200789296,
|
| 14 |
+
"grad_norm": 7.956792831420898,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.8474,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 5.635889530181885,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.5684,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 1.822274923324585,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.1416,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.40320682525634766,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9632,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.2249278426170349,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.934,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.1623561829328537,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.8913,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.12772700190544128,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.8674,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.1415039449930191,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.8754,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.1315464973449707,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.8631,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.12670297920703888,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8399,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.1190510168671608,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8266,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.12235772609710693,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8368,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.11031907796859741,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8258,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.12731650471687317,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8281,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.14221690595149994,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.8208,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.12399591505527496,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.819,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.10620061308145523,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8165,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.12593528628349304,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8419,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.12120434641838074,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8414,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.12849418818950653,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8253,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.11673939973115921,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8261,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.1167469322681427,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8059,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.13127610087394714,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.815,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.1274745911359787,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.8013,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.1188514307141304,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.815,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.13758516311645508,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8212,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.11778578162193298,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.8074,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.11159393936395645,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8128,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.11813396960496902,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8264,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.11451815068721771,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.7866,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.11813846230506897,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.8005,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.11911551654338837,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.7998,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.12456249445676804,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.8009,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.11578356474637985,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.8126,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.12272408604621887,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.8163,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.11508162319660187,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.7994,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.11359044164419174,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8225,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.11524739116430283,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.7959,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.11921390891075134,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.8006,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.10500162839889526,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.8045,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.11149939894676208,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.7893,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.11901786923408508,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.7994,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.11487217992544174,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.7928,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.12109464406967163,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.7969,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.11583343148231506,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7781,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.11327127367258072,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.7935,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.11830000579357147,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.7987,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.11164557933807373,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.7939,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.1029757633805275,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7833,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.106364905834198,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.7884,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.11867601424455643,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7713,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.11959987133741379,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.8123,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.12902922928333282,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.7823,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.12428376823663712,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.7982,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.11284665018320084,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.8023,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.11135192960500717,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.8043,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.11998610943555832,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8178,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.11443104594945908,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.7875,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.11247782409191132,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.8081,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.12725119292736053,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.8044,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.12547408044338226,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.8134,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.11902306228876114,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.8055,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.1228843554854393,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8115,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.1080876812338829,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.8044,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.10257679224014282,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7604,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.12421458959579468,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.7926,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.11707786470651627,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.8041,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.110695019364357,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7829,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.11130692064762115,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.7962,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.11394365131855011,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.7996,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.11141709238290787,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7847,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.10854491591453552,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.7888,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.11427657306194305,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.7989,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.12043122947216034,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.7819,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.10522384941577911,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.7807,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.10944320261478424,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7733,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.11377627402544022,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.7913,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.11907906830310822,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.7847,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.11925205588340759,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.7766,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.11634136736392975,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.8031,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.11499983072280884,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.7901,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.10716079920530319,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.7932,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.11633600294589996,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7798,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.12170834839344025,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7824,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.1074165552854538,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.7833,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.10895487666130066,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.78,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.11056454479694366,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.7769,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.10637789219617844,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7828,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.10883282124996185,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.7734,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.10449407249689102,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.7795,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.11820509284734726,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.7759,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.1080690324306488,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.764,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.12423931807279587,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7617,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.1336292326450348,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.796,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.11137071996927261,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8057,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.12131278216838837,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.7864,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.11296875029802322,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7674,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.10804447531700134,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7748,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.1191755011677742,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.763,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.10823065042495728,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.7826,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.10970450192689896,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.8021,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.11690064519643784,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.7832,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.10771074146032333,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.7939,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.1085444986820221,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.7927,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.10309944301843643,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.7846,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.11519423127174377,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.7891,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.11497539281845093,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7696,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.11025021970272064,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.8032,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.12225687503814697,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.7785,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.11960980296134949,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.8,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.11596061289310455,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.7946,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.1274496465921402,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.7889,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.11255494505167007,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7759,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.11006594449281693,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.7799,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.11827494204044342,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.779,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.1290554702281952,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.8025,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.10886631906032562,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8132,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.10158252716064453,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.7821,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.11702360212802887,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7667,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.118947833776474,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7741,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.11493363231420517,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7778,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.12264607101678848,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.7925,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.10500648617744446,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.7961,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.11112629622220993,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7718,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.11084358394145966,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7724,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.11696505546569824,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7731,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.12438218295574188,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7612,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.10892562568187714,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7681,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.11126788705587387,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7877,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.11376596242189407,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.7831,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.11741607636213303,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.7936,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.12631665170192719,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7755,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.14833082258701324,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7769,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.11646168678998947,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7794,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.11347928643226624,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7693,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.13203497231006622,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.785,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.11332967132329941,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.7917,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.10165880620479584,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.7934,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.12402689456939697,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7726,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.11317344754934311,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7774,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.11599227041006088,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.7922,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.11761463433504105,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7793,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.11712560802698135,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.789,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.11572546511888504,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.7916,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.11936511099338531,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.7724,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.11254958808422089,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7782,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.1100882813334465,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7743,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.11968721449375153,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7728,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.11492651700973511,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7801,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.11040227115154266,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.7872,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.11208613216876984,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.7794,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.11467964202165604,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.7888,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.11829792708158493,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.7874,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.10884705185890198,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.7937,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.11565709859132767,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.7872,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.12199011445045471,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.8013,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.11875002086162567,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.7882,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.11751268059015274,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.7875,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.10103359073400497,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.7854,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.10703160613775253,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7696,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.10172867774963379,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7513,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.11436061561107635,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7562,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.11003131419420242,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.7913,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.09882090240716934,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7723,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.11715419590473175,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.7842,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.11415360122919083,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7693,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.1219923123717308,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.788,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.12011914700269699,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.7773,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.11282134056091309,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7656,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.10229738801717758,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7662,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.11370555311441422,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.7824,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.11261133849620819,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.772,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.11336389183998108,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7784,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.11519230902194977,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7813,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.11225500702857971,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7788,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.1033378541469574,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7675,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.11400485783815384,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8035,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.11835762858390808,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.7887,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.1097395047545433,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.801,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.10827604681253433,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7659,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.1143343448638916,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7626,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.10988623648881912,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.7844,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.10689529776573181,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7795,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.1233961284160614,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7807,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.12096843868494034,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.7865,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.11251448094844818,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.786,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.10431496798992157,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7795,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.1128111332654953,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.7851,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.11870326101779938,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7704,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.11834660172462463,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.7825,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.10816121846437454,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.784,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.11076580733060837,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7738,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.11351194977760315,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.79,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.11363471299409866,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.7965,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.11174146085977554,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.7802,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.11743441224098206,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.8001,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.11296823620796204,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.7914,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.11472656577825546,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.7816,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.1253628432750702,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.7788,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.10886093974113464,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7798,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.12469318509101868,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7796,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.10476791113615036,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7655,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.11816437542438507,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7566,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.11472361534833908,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7726,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.10908802598714828,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7768,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.10618213564157486,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7613,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.12522989511489868,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.7768,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.10823426395654678,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7597,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.10466082394123077,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.7846,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.11098673194646835,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.778,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.11172748357057571,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.7864,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.11382188647985458,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.7844,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.11567959189414978,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7709,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.11197977513074875,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.7819,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.13054481148719788,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7698,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.11312831938266754,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.7838,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.11173035949468613,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.7889,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.10865781456232071,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7739,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.10864192992448807,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.7887,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.12094397842884064,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7708,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.12824127078056335,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7561,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.10331481695175171,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.7808,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.11992459744215012,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7723,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.11258258670568466,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.7894,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.11250903457403183,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.776,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.11421409249305725,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7641,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.10767006874084473,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7686,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.11010268330574036,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7791,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.10855183750391006,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.7872,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.10499639064073563,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7696,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.10421514511108398,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7735,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.10575596988201141,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.7797,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.11518104374408722,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.79,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.10762431472539902,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.7822,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.11134110391139984,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7633,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.11296787112951279,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7815,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.1162649616599083,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.766,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.11281555145978928,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7728,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.1090969666838646,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.7882,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.11857166141271591,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.7792,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.10877195000648499,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7658,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.11024946719408035,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.7934,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.11875810474157333,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.7863,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.1161121279001236,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.776,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.12308669835329056,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.777,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.11371367424726486,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.7922,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.11282896995544434,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.7808,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.1118343323469162,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7737,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.1104198545217514,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.7776,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.11186999082565308,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7703,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.11136513948440552,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7676,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.10684780031442642,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.78,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.11467868834733963,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7724,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.10620006918907166,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7738,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.10817451775074005,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.7849,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.09870022535324097,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7668,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.11812776327133179,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7699,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.1106356531381607,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7579,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.10907068103551865,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.782,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.877332083881804e+18,
|
| 1829 |
+
"train_loss": 0.7976896363494486,
|
| 1830 |
+
"train_runtime": 10614.2051,
|
| 1831 |
+
"train_samples_per_second": 62.643,
|
| 1832 |
+
"train_steps_per_second": 0.245
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 332174164,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 3.877332083881804e+18,
|
| 1853 |
+
"train_batch_size": 4,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/ijepa_vith14/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68657a8a3b741a30b098c2879827bca213715d72fef25cf892b790023ce6cabc
|
| 3 |
+
size 6264
|
continuous/qwen25/ijepa_vith14/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/ijepa_vith14/pretrain/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen2ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"layer_types": [
|
| 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 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"max_position_embeddings": 32768,
|
| 51 |
+
"max_window_layers": 21,
|
| 52 |
+
"mm_hidden_size": 1280,
|
| 53 |
+
"mm_newline_position": "grid",
|
| 54 |
+
"mm_patch_merge_type": "flat",
|
| 55 |
+
"mm_projector_lr": null,
|
| 56 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 57 |
+
"mm_resampler_type": null,
|
| 58 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 59 |
+
"mm_spatial_pool_stride": null,
|
| 60 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 61 |
+
"mm_use_im_patch_token": false,
|
| 62 |
+
"mm_use_im_start_end": false,
|
| 63 |
+
"mm_vision_select_feature": "patch",
|
| 64 |
+
"mm_vision_select_layer": -1,
|
| 65 |
+
"mm_vision_tower": "vtb_ssl:ijepa",
|
| 66 |
+
"mm_vision_tower_lr": null,
|
| 67 |
+
"model_type": "qwen2",
|
| 68 |
+
"num_attention_heads": 12,
|
| 69 |
+
"num_hidden_layers": 28,
|
| 70 |
+
"num_key_value_heads": 2,
|
| 71 |
+
"pad_token_id": 151643,
|
| 72 |
+
"pos_skipping_range": 4096,
|
| 73 |
+
"rms_norm_eps": 1e-06,
|
| 74 |
+
"rope_scaling": null,
|
| 75 |
+
"rope_theta": 1000000.0,
|
| 76 |
+
"sliding_window": null,
|
| 77 |
+
"tie_word_embeddings": true,
|
| 78 |
+
"tokenizer_model_max_length": 8192,
|
| 79 |
+
"tokenizer_padding_side": "right",
|
| 80 |
+
"transformers_version": "4.57.6",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"use_mm_proj": true,
|
| 83 |
+
"use_pos_skipping": false,
|
| 84 |
+
"use_sliding_window": false,
|
| 85 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/ijepa/ijepa_vith.pth",
|
| 86 |
+
"vocab_size": 151936
|
| 87 |
+
}
|
continuous/qwen25/ijepa_vith14/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4332a7896feb59b1f841c671da1a20eef1ad2799da35ddb3390c10f952cf654
|
| 3 |
+
size 17315896
|
continuous/qwen25/ijepa_vith14/pretrain/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 545,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01834862385321101,
|
| 14 |
+
"grad_norm": 0.17368356883525848,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.862,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.03070809692144394,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 4.1113,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.03663262724876404,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.8352,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.0236353762447834,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.4148,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.01297892164438963,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.2966,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.011160054244101048,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.2402,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.011949322186410427,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.2011,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.012162646278738976,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 3.1746,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.013932217843830585,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 3.1431,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.01641245372593403,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 3.1315,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.013708814978599548,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 3.1102,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.02013043314218521,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 3.1032,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.020167594775557518,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 3.081,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.01816217228770256,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 3.0713,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.02121996507048607,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 3.0693,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.014565208926796913,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 3.0569,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.017308179289102554,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 3.0537,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.020073411986231804,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 3.0417,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.015896789729595184,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 3.0403,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.015621492639183998,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 3.0208,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.02440403215587139,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 3.0178,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.016779348254203796,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 3.0257,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.018784228712320328,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 3.0275,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.018059104681015015,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 3.014,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.016315028071403503,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 3.001,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.016537386924028397,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 3.0142,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.01733884960412979,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.9949,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.015220416709780693,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.9971,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.014501553028821945,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.9805,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.01534369308501482,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.9831,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.017069393768906593,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.9913,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.016626551747322083,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.9866,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.016483241692185402,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.9717,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.015483876690268517,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.9755,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.016719508916139603,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.9823,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.02016097865998745,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.9809,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.016243955120444298,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.9717,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.015011302195489407,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.9806,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.01948518119752407,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.9726,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.014385068789124489,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.9672,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.013658720068633556,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.9641,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.014039728790521622,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.962,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.014534947462379932,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.9661,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.0166236013174057,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.9728,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.01435981784015894,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.9682,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.013529621995985508,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.96,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.01459386758506298,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.9482,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.014213300310075283,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.9575,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.014948908239603043,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.9622,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.013363175094127655,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.9689,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.014791174791753292,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.9596,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.013464661315083504,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.9541,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.013607091270387173,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.9566,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.013133345171809196,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.9589,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 3.408933495798497e+17,
|
| 394 |
+
"train_loss": 3.09800087815031,
|
| 395 |
+
"train_runtime": 5296.9132,
|
| 396 |
+
"train_samples_per_second": 105.369,
|
| 397 |
+
"train_steps_per_second": 0.103
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 29204608,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 3.408933495798497e+17,
|
| 418 |
+
"train_batch_size": 8,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/mc1_b16_224_2.5b/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</tool_call>": 151658,
|
| 3 |
+
"<tool_call>": 151657,
|
| 4 |
+
"<|box_end|>": 151649,
|
| 5 |
+
"<|box_start|>": 151648,
|
| 6 |
+
"<|endoftext|>": 151643,
|
| 7 |
+
"<|file_sep|>": 151664,
|
| 8 |
+
"<|fim_middle|>": 151660,
|
| 9 |
+
"<|fim_pad|>": 151662,
|
| 10 |
+
"<|fim_prefix|>": 151659,
|
| 11 |
+
"<|fim_suffix|>": 151661,
|
| 12 |
+
"<|im_end|>": 151645,
|
| 13 |
+
"<|im_start|>": 151644,
|
| 14 |
+
"<|image_pad|>": 151655,
|
| 15 |
+
"<|object_ref_end|>": 151647,
|
| 16 |
+
"<|object_ref_start|>": 151646,
|
| 17 |
+
"<|quad_end|>": 151651,
|
| 18 |
+
"<|quad_start|>": 151650,
|
| 19 |
+
"<|repo_name|>": 151663,
|
| 20 |
+
"<|video_pad|>": 151656,
|
| 21 |
+
"<|vision_end|>": 151653,
|
| 22 |
+
"<|vision_pad|>": 151654,
|
| 23 |
+
"<|vision_start|>": 151652
|
| 24 |
+
}
|
continuous/qwen25/mc1_b16_224_2.5b/finetune/config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwenForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 151643,
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"faster_token_stride": 10,
|
| 11 |
+
"force_sample": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 1536,
|
| 14 |
+
"image_aspect_ratio": "square",
|
| 15 |
+
"image_crop_resolution": null,
|
| 16 |
+
"image_grid_pinpoints": null,
|
| 17 |
+
"image_split_resolution": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 8960,
|
| 20 |
+
"max_position_embeddings": 32768,
|
| 21 |
+
"max_window_layers": 21,
|
| 22 |
+
"mm_hidden_size": 768,
|
| 23 |
+
"mm_newline_position": "grid",
|
| 24 |
+
"mm_patch_merge_type": "flat",
|
| 25 |
+
"mm_projector_lr": 0.0001,
|
| 26 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 27 |
+
"mm_resampler_type": null,
|
| 28 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 29 |
+
"mm_spatial_pool_stride": null,
|
| 30 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 31 |
+
"mm_use_im_patch_token": false,
|
| 32 |
+
"mm_use_im_start_end": false,
|
| 33 |
+
"mm_vision_select_feature": "patch",
|
| 34 |
+
"mm_vision_select_layer": -2,
|
| 35 |
+
"mm_vision_tower": "open_clip_hub:ViT-B-16",
|
| 36 |
+
"mm_vision_tower_lr": null,
|
| 37 |
+
"model_type": "qwen2",
|
| 38 |
+
"num_attention_heads": 12,
|
| 39 |
+
"num_hidden_layers": 28,
|
| 40 |
+
"num_key_value_heads": 2,
|
| 41 |
+
"pos_skipping_range": 4096,
|
| 42 |
+
"rms_norm_eps": 1e-06,
|
| 43 |
+
"rope_scaling": null,
|
| 44 |
+
"rope_theta": 1000000.0,
|
| 45 |
+
"sliding_window": 32768,
|
| 46 |
+
"tie_word_embeddings": true,
|
| 47 |
+
"tokenizer_model_max_length": 8192,
|
| 48 |
+
"tokenizer_padding_side": "right",
|
| 49 |
+
"torch_dtype": "bfloat16",
|
| 50 |
+
"transformers_version": "4.51.3",
|
| 51 |
+
"use_cache": true,
|
| 52 |
+
"use_mm_proj": true,
|
| 53 |
+
"use_pos_skipping": false,
|
| 54 |
+
"use_sliding_window": false,
|
| 55 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/mc1/mc1_b16_224_2.5b.pt",
|
| 56 |
+
"vocab_size": 151936
|
| 57 |
+
}
|