Upload continuous evaluation checkpoints, batch 5
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +3 -0
- continuous/qwen25/mc2_b16_384/pretrain/config.json +58 -0
- continuous/qwen25/mc2_b16_384/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/mc2_b16_384/pretrain/trainer_state.json +421 -0
- continuous/qwen25/mc2_b32_224/finetune/added_tokens.json +24 -0
- continuous/qwen25/mc2_b32_224/finetune/config.json +57 -0
- continuous/qwen25/mc2_b32_224/finetune/generation_config.json +14 -0
- continuous/qwen25/mc2_b32_224/finetune/merges.txt +0 -0
- continuous/qwen25/mc2_b32_224/finetune/model.safetensors +3 -0
- continuous/qwen25/mc2_b32_224/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/mc2_b32_224/finetune/tokenizer.json +3 -0
- continuous/qwen25/mc2_b32_224/finetune/tokenizer_config.json +209 -0
- continuous/qwen25/mc2_b32_224/finetune/trainer_state.json +1856 -0
- continuous/qwen25/mc2_b32_224/finetune/training_args.bin +3 -0
- continuous/qwen25/mc2_b32_224/finetune/vocab.json +0 -0
- continuous/qwen25/mc2_b32_224/pretrain/config.json +58 -0
- continuous/qwen25/mc2_b32_224/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/mc2_b32_224/pretrain/trainer_state.json +421 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/added_tokens.json +24 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/config.json +57 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/generation_config.json +14 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/merges.txt +0 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/model.safetensors +3 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/tokenizer.json +3 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/tokenizer_config.json +209 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/trainer_state.json +1856 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/training_args.bin +3 -0
- continuous/qwen25/mc2_b32_224_mt5/finetune/vocab.json +0 -0
- continuous/qwen25/mc2_b32_224_mt5/pretrain/config.json +58 -0
- continuous/qwen25/mc2_b32_224_mt5/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/mc2_b32_224_mt5/pretrain/trainer_state.json +421 -0
- continuous/qwen25/mc2_b32_384/finetune/added_tokens.json +24 -0
- continuous/qwen25/mc2_b32_384/finetune/config.json +57 -0
- continuous/qwen25/mc2_b32_384/finetune/generation_config.json +14 -0
- continuous/qwen25/mc2_b32_384/finetune/merges.txt +0 -0
- continuous/qwen25/mc2_b32_384/finetune/model.safetensors +3 -0
- continuous/qwen25/mc2_b32_384/finetune/special_tokens_map.json +31 -0
- continuous/qwen25/mc2_b32_384/finetune/tokenizer.json +3 -0
- continuous/qwen25/mc2_b32_384/finetune/tokenizer_config.json +209 -0
- continuous/qwen25/mc2_b32_384/finetune/trainer_state.json +1856 -0
- continuous/qwen25/mc2_b32_384/finetune/training_args.bin +3 -0
- continuous/qwen25/mc2_b32_384/finetune/vocab.json +0 -0
- continuous/qwen25/mc2_b32_384/pretrain/config.json +58 -0
- continuous/qwen25/mc2_b32_384/pretrain/mm_projector.bin +3 -0
- continuous/qwen25/mc2_b32_384/pretrain/trainer_state.json +421 -0
- continuous/qwen25/mc2_g14_224/finetune/added_tokens.json +24 -0
- continuous/qwen25/mc2_g14_224/finetune/config.json +57 -0
- continuous/qwen25/mc2_g14_224/finetune/generation_config.json +14 -0
- continuous/qwen25/mc2_g14_224/finetune/merges.txt +0 -0
.gitattributes
CHANGED
|
@@ -62,3 +62,6 @@ continuous/qwen25/mc1_l14_224_2.5b/finetune/tokenizer.json filter=lfs diff=lfs m
|
|
| 62 |
continuous/qwen25/mc1_l14_224_400m/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 63 |
continuous/qwen25/mc2_b16_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 64 |
continuous/qwen25/mc2_b16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
continuous/qwen25/mc1_l14_224_400m/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 63 |
continuous/qwen25/mc2_b16_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 64 |
continuous/qwen25/mc2_b16_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
continuous/qwen25/mc2_b32_224/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
continuous/qwen25/mc2_b32_224_mt5/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
continuous/qwen25/mc2_b32_384/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
continuous/qwen25/mc2_b16_384/pretrain/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"add_faster_video": false,
|
| 4 |
+
"add_time_instruction": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"bos_token_id": 151643,
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 1536,
|
| 15 |
+
"image_aspect_ratio": "square",
|
| 16 |
+
"image_crop_resolution": null,
|
| 17 |
+
"image_grid_pinpoints": null,
|
| 18 |
+
"image_split_resolution": null,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 8960,
|
| 21 |
+
"max_position_embeddings": 32768,
|
| 22 |
+
"max_window_layers": 21,
|
| 23 |
+
"mm_hidden_size": 768,
|
| 24 |
+
"mm_newline_position": "grid",
|
| 25 |
+
"mm_patch_merge_type": "flat",
|
| 26 |
+
"mm_projector_lr": null,
|
| 27 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 28 |
+
"mm_resampler_type": null,
|
| 29 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 30 |
+
"mm_spatial_pool_stride": null,
|
| 31 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 32 |
+
"mm_use_im_patch_token": false,
|
| 33 |
+
"mm_use_im_start_end": false,
|
| 34 |
+
"mm_vision_select_feature": "patch",
|
| 35 |
+
"mm_vision_select_layer": -2,
|
| 36 |
+
"mm_vision_tower": "open_clip_hub:ViT-B-16",
|
| 37 |
+
"mm_vision_tower_lr": null,
|
| 38 |
+
"model_type": "qwen2",
|
| 39 |
+
"num_attention_heads": 12,
|
| 40 |
+
"num_hidden_layers": 28,
|
| 41 |
+
"num_key_value_heads": 2,
|
| 42 |
+
"pos_skipping_range": 4096,
|
| 43 |
+
"rms_norm_eps": 1e-06,
|
| 44 |
+
"rope_scaling": null,
|
| 45 |
+
"rope_theta": 1000000.0,
|
| 46 |
+
"sliding_window": 32768,
|
| 47 |
+
"tie_word_embeddings": true,
|
| 48 |
+
"tokenizer_model_max_length": 8192,
|
| 49 |
+
"tokenizer_padding_side": "right",
|
| 50 |
+
"torch_dtype": "bfloat16",
|
| 51 |
+
"transformers_version": "4.51.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"use_mm_proj": true,
|
| 54 |
+
"use_pos_skipping": false,
|
| 55 |
+
"use_sliding_window": false,
|
| 56 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/mc2/mc2_b16_384.pt",
|
| 57 |
+
"vocab_size": 151936
|
| 58 |
+
}
|
continuous/qwen25/mc2_b16_384/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ade452eddce7968639eee96113eddfb3328b92d972e066f856707cbcda21cfa8
|
| 3 |
+
size 14170168
|
continuous/qwen25/mc2_b16_384/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": 1.3795558214187622,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.7194,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.2047562450170517,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.5876,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.0675281435251236,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.3131,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.036457862704992294,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.2554,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.04061898589134216,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.2357,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.039856381714344025,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.2102,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.06858548521995544,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.1872,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.06435181945562363,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 3.163,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.05282606929540634,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 3.1233,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.0686253160238266,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 3.096,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.0724925845861435,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 3.0581,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.07010164111852646,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 3.0319,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.061458658427000046,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 2.9855,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.05375082790851593,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 2.9612,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.07433775067329407,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 2.9436,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.06720367819070816,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 2.9128,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.06241718679666519,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.8991,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.08703015744686127,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.8792,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.10212589055299759,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.8673,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.06709428131580353,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.8413,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.06795889139175415,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.8315,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.07126250863075256,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.8329,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.06353778392076492,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.8221,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.06900804489850998,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.8046,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.05861612409353256,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.7832,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.060657575726509094,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.7816,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.0600074902176857,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.7474,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.07425099611282349,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.7356,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.07085687667131424,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.7064,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.06921140104532242,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.6909,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.08273240923881531,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.6882,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.06653282791376114,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.6741,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.09017304331064224,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.6539,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.06981606781482697,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.6497,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.064715176820755,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.6484,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.08981294184923172,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.645,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.07130980491638184,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.6337,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.06873934715986252,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.6404,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.06646282970905304,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.6305,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.0661286786198616,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.6178,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.05849597230553627,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.6203,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.05707545951008797,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.6148,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.06218673661351204,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.6121,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.06620001792907715,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.6225,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.06459390372037888,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.6134,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.0593554712831974,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.6058,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.060893308371305466,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.5981,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.06663157045841217,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.5996,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.061391811817884445,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.6092,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.060457002371549606,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.6159,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.061244916170835495,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.6043,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.06044212356209755,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.6033,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.05542055889964104,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.6047,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.061167895793914795,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.6012,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.6800122460582707e+17,
|
| 394 |
+
"train_loss": 2.8499282346952945,
|
| 395 |
+
"train_runtime": 5404.35,
|
| 396 |
+
"train_samples_per_second": 103.274,
|
| 397 |
+
"train_steps_per_second": 0.101
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 0,
|
| 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": 2.6800122460582707e+17,
|
| 418 |
+
"train_batch_size": 32,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/mc2_b32_224/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/mc2_b32_224/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-32",
|
| 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/mc2/mc2_b32_224.pt",
|
| 56 |
+
"vocab_size": 151936
|
| 57 |
+
}
|
continuous/qwen25/mc2_b32_224/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.51.3"
|
| 14 |
+
}
|
continuous/qwen25/mc2_b32_224/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_224/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91d8ddf38eb85e4b0e6584e5af926be2d68baed4a3430f230b8fcc58daa51b22
|
| 3 |
+
size 3744103520
|
continuous/qwen25/mc2_b32_224/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/mc2_b32_224/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/mc2_b32_224/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"extra_special_tokens": {},
|
| 203 |
+
"model_max_length": 8192,
|
| 204 |
+
"pad_token": "<|endoftext|>",
|
| 205 |
+
"padding_side": "right",
|
| 206 |
+
"split_special_tokens": false,
|
| 207 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 208 |
+
"unk_token": null
|
| 209 |
+
}
|
continuous/qwen25/mc2_b32_224/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": 17.200342178344727,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.9681,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 6.99927282333374,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.6234,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 3.5848217010498047,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.1835,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.9100845456123352,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9761,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.36845239996910095,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.9292,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.28475892543792725,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.8774,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.2256624847650528,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.8489,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.2375335544347763,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.8567,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.21229559183120728,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.8372,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.2322784960269928,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8167,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.21149080991744995,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.806,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.22437211871147156,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8177,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.206223264336586,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8082,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.23353156447410583,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8093,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.23030319809913635,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.7918,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.21488964557647705,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8001,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.19619038701057434,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8007,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.22447940707206726,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8266,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.22034280002117157,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8178,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.23714692890644073,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8107,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.21100960671901703,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.802,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.21457965672016144,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.789,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.21754753589630127,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.7919,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.2250698357820511,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.7852,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.22228537499904633,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8022,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.21462349593639374,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.20451626181602478,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.7893,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.19934901595115662,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.7884,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.21576984226703644,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8085,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.19738611578941345,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.7705,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.21261398494243622,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.7793,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.21530015766620636,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.7845,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.22078928351402283,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.7821,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.20841793715953827,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.7951,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.22101925313472748,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.7942,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.21127699315547943,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.7807,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.21001914143562317,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8057,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.21378497779369354,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.7713,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.21553750336170197,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.7732,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.19726921617984772,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.7792,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.20711632072925568,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.7707,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.21608886122703552,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.7853,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.20983542501926422,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.7779,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.201375812292099,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.7753,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.2134191244840622,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7645,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.2137448936700821,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.7703,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.20202553272247314,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.776,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.21203021705150604,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.7761,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.19458362460136414,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7619,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.19448433816432953,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.7727,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.2163534164428711,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7632,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.22652246057987213,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.7932,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.21960335969924927,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.7667,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.22350087761878967,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.7786,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.2055044323205948,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.7815,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.20114873349666595,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.7839,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.22615838050842285,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.7948,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.22480538487434387,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.7732,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.21647901833057404,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.7855,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.2159341424703598,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.7759,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.22373400628566742,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.7969,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.20451626181602478,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.7802,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.2220970243215561,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.7964,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.2043190449476242,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.7876,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.19304001331329346,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.746,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.22540433704853058,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.7785,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.23418457806110382,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.785,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.2010469138622284,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7544,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.21348454058170319,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.7763,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.21255528926849365,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.7832,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.20971277356147766,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7659,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.197662353515625,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.7696,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.20884989202022552,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.7738,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.21702146530151367,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.7635,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.1997770220041275,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.7626,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.2080141305923462,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7553,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.20480592548847198,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.7731,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.20119036734104156,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.7691,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.22222265601158142,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.7562,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.20293574035167694,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.7835,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.21633483469486237,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.7755,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.1967669427394867,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.7697,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.20819856226444244,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7547,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.22472181916236877,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7603,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.21377217769622803,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.772,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.20802189409732819,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7569,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.20660437643527985,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.7662,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.199713796377182,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7617,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.1977252960205078,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.7607,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.18925337493419647,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.7649,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.21756452322006226,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.75,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.2227257341146469,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7532,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.2194576859474182,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7408,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.22842873632907867,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.7746,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.21674393117427826,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.7884,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.22588028013706207,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.7748,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.21190889179706573,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7542,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.20660613477230072,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7601,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.2175944447517395,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7429,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.20466063916683197,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.76,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.2062431275844574,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.7851,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.2197045236825943,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.7768,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.20419752597808838,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.7737,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.21289226412773132,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.7641,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.1952154040336609,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.7607,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.19230562448501587,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.7663,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.21263642609119415,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7568,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.20245547592639923,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.781,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.2303694784641266,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.7642,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.20842845737934113,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.7782,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2136482149362564,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.7744,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.22622261941432953,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.7698,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.20889048278331757,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7476,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.20704306662082672,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.7697,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.21975214779376984,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.7615,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.24042904376983643,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.7848,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.21143706142902374,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.7958,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.19593015313148499,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.7688,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.21661891043186188,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.748,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.2163098007440567,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7613,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.21622903645038605,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7626,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.21462152898311615,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.78,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.2046796679496765,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.7839,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.2037946581840515,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7521,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.19872234761714935,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.751,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2223239243030548,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7599,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.23529595136642456,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7422,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.21097756922245026,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7513,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.21381385624408722,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7618,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.19897615909576416,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.7641,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.19925685226917267,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.7761,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.2127237468957901,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7473,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.2256951481103897,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7551,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.21414275467395782,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7622,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.21467985212802887,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7557,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.24019847810268402,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.7732,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.21698299050331116,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.7695,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.19704867899417877,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.7776,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.2251235991716385,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7576,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.20770388841629028,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7574,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.21898683905601501,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.7708,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.22126980125904083,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7551,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.2237810343503952,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.7712,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.2183912843465805,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.7756,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.21837463974952698,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.7555,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.21747155487537384,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7622,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.2060493677854538,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7577,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.21733126044273376,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7541,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.2021649032831192,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7608,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.19979721307754517,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.7701,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.21114248037338257,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.7636,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.2081618756055832,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.7717,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.20986056327819824,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.7721,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.20802699029445648,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.7779,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.21284809708595276,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.764,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.20923694968223572,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.7788,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.21899579465389252,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.7758,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.22611485421657562,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.7654,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.18582072854042053,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.768,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.1997162103652954,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7556,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.19890141487121582,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7377,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.21137970685958862,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7372,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.20957571268081665,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.7691,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.186448872089386,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7582,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.2130914032459259,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.7685,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.20848390460014343,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7484,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.2198670655488968,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.7688,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.2144138514995575,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.7647,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.2189735323190689,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7547,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.1897955983877182,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7537,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.21521498262882233,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.7663,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.21669965982437134,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7557,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.21084590256214142,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7611,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.2142496556043625,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7611,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.1995730996131897,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7595,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.20149825513362885,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7475,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.22077330946922302,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.7907,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.22013911604881287,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.7758,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.20601831376552582,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.787,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.207325279712677,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7492,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.22383329272270203,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7429,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.20376676321029663,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.7638,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.20204110443592072,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7541,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.22023743391036987,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7614,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.21888136863708496,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.7709,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.2130376398563385,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.7657,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.19645746052265167,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7614,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.2027592658996582,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.7697,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.21927107870578766,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7522,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.21453061699867249,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.7625,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.20093508064746857,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.76,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.20657506585121155,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7542,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.2164606899023056,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.7688,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.21846213936805725,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.7757,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.21379496157169342,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.7635,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.22881406545639038,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.7815,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.2077098786830902,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.7708,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.22078150510787964,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.7716,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.21955548226833344,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.7626,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.2071690410375595,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7555,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.20900270342826843,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7644,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.19893930852413177,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7517,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.20551688969135284,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7367,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.20705927908420563,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7531,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.20012661814689636,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7577,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.19430120289325714,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7416,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.22419032454490662,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.7635,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.19466128945350647,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7483,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.19959035515785217,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.7681,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.20534047484397888,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7568,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.20770332217216492,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.7665,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.21190133690834045,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.7677,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.22288601100444794,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7523,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.21048712730407715,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.758,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.22758011519908905,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7489,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.20131905376911163,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.7677,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.20718242228031158,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.7681,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.20405903458595276,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7589,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.20047815144062042,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.7664,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.2186414897441864,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7614,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.21161913871765137,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7314,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.19474974274635315,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.7647,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.21021072566509247,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7604,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.19382567703723907,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.7711,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.19865897297859192,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.756,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.2106754332780838,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7457,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.20593629777431488,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7537,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.20924821496009827,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7638,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.20984616875648499,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.7682,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.19534173607826233,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7468,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.19388101994991302,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7594,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.20452211797237396,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.7625,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.20420224964618683,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.7674,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.20408868789672852,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.7607,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.20560859143733978,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7455,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.20563532412052155,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7571,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.22346307337284088,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7507,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.207896426320076,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7592,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.20672740042209625,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.77,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.22376953065395355,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.7697,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.20797251164913177,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7502,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.2086784988641739,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.7755,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.21670398116111755,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.7693,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.2184397280216217,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.7594,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.22616046667099,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7564,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.20686067640781403,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.776,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.22115574777126312,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.7711,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.20681682229042053,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7522,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.21081092953681946,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.759,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.2071520835161209,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7558,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.20704740285873413,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7483,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.21442711353302002,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.7687,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.20767924189567566,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7552,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.19858328998088837,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7563,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.19989663362503052,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.7651,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.20718428492546082,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7572,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.20673608779907227,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7479,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.2025166153907776,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7397,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.2122281789779663,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.7717,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.522737300075184e+18,
|
| 1829 |
+
"train_loss": 0.7804976121067404,
|
| 1830 |
+
"train_runtime": 9279.8837,
|
| 1831 |
+
"train_samples_per_second": 71.65,
|
| 1832 |
+
"train_steps_per_second": 0.28
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 0,
|
| 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": 3.522737300075184e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/mc2_b32_224/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11855b40de94ce084eeda081cc8f5ea79dcf4b879cdcde01036b9fdfd19607e3
|
| 3 |
+
size 6136
|
continuous/qwen25/mc2_b32_224/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_224/pretrain/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"add_faster_video": false,
|
| 4 |
+
"add_time_instruction": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"bos_token_id": 151643,
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 1536,
|
| 15 |
+
"image_aspect_ratio": "square",
|
| 16 |
+
"image_crop_resolution": null,
|
| 17 |
+
"image_grid_pinpoints": null,
|
| 18 |
+
"image_split_resolution": null,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 8960,
|
| 21 |
+
"max_position_embeddings": 32768,
|
| 22 |
+
"max_window_layers": 21,
|
| 23 |
+
"mm_hidden_size": 768,
|
| 24 |
+
"mm_newline_position": "grid",
|
| 25 |
+
"mm_patch_merge_type": "flat",
|
| 26 |
+
"mm_projector_lr": null,
|
| 27 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 28 |
+
"mm_resampler_type": null,
|
| 29 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 30 |
+
"mm_spatial_pool_stride": null,
|
| 31 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 32 |
+
"mm_use_im_patch_token": false,
|
| 33 |
+
"mm_use_im_start_end": false,
|
| 34 |
+
"mm_vision_select_feature": "patch",
|
| 35 |
+
"mm_vision_select_layer": -2,
|
| 36 |
+
"mm_vision_tower": "open_clip_hub:ViT-B-32",
|
| 37 |
+
"mm_vision_tower_lr": null,
|
| 38 |
+
"model_type": "qwen2",
|
| 39 |
+
"num_attention_heads": 12,
|
| 40 |
+
"num_hidden_layers": 28,
|
| 41 |
+
"num_key_value_heads": 2,
|
| 42 |
+
"pos_skipping_range": 4096,
|
| 43 |
+
"rms_norm_eps": 1e-06,
|
| 44 |
+
"rope_scaling": null,
|
| 45 |
+
"rope_theta": 1000000.0,
|
| 46 |
+
"sliding_window": 32768,
|
| 47 |
+
"tie_word_embeddings": true,
|
| 48 |
+
"tokenizer_model_max_length": 8192,
|
| 49 |
+
"tokenizer_padding_side": "right",
|
| 50 |
+
"torch_dtype": "bfloat16",
|
| 51 |
+
"transformers_version": "4.51.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"use_mm_proj": true,
|
| 54 |
+
"use_pos_skipping": false,
|
| 55 |
+
"use_sliding_window": false,
|
| 56 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/mc2/mc2_b32_224.pt",
|
| 57 |
+
"vocab_size": 151936
|
| 58 |
+
}
|
continuous/qwen25/mc2_b32_224/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cee29ac56157d2d5713c4414d6994984e8c5dffe7173b55f0ce8b5837776f89d
|
| 3 |
+
size 14170168
|
continuous/qwen25/mc2_b32_224/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": 2.303893566131592,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.3413,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.5821881294250488,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.4062,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.1756521761417389,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.2492,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.2801302969455719,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.1644,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.1970607489347458,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.1069,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.1643131524324417,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.0545,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.2219686061143875,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 3.013,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.2654235064983368,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 2.9807,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.1918623000383377,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 2.948,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.20921580493450165,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 2.926,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.20447315275669098,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 2.9042,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.20269811153411865,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 2.896,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.2591926157474518,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 2.866,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.22893363237380981,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 2.856,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.2664387822151184,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 2.8535,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.22189420461654663,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 2.8344,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.2098989486694336,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.8214,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.2496766597032547,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.8018,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.22642163932323456,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.7847,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.21990275382995605,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.7558,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.23548953235149384,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.7431,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.21344129741191864,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.7489,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.20757056772708893,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.7377,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.2072771191596985,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.7228,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.19506749510765076,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.7143,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.20137012004852295,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.7227,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.2099602222442627,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.7029,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.24027952551841736,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.7064,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.19804129004478455,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.6919,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.20738054811954498,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.6905,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.22681790590286255,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.6998,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.1971491128206253,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.6937,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.22741390764713287,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.6782,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.293171763420105,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.6847,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.1964290887117386,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.6833,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.22073006629943848,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.686,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.18965387344360352,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.6794,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.1847037971019745,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.6846,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.20089305937290192,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.6736,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.18381892144680023,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.6682,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.21136370301246643,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.6692,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.18057624995708466,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.6697,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.1676991581916809,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.6653,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.18617069721221924,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.6726,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.17468911409378052,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.6691,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.18463601171970367,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.661,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.18165278434753418,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.6524,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.18050505220890045,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.6576,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.17650410532951355,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.6664,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.1794212907552719,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.6709,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.17906996607780457,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.6575,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.17159675061702728,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.6575,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.1690247654914856,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.6595,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.18075332045555115,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.6583,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.8922594581282816e+17,
|
| 394 |
+
"train_loss": 2.8111128535839396,
|
| 395 |
+
"train_runtime": 5304.8952,
|
| 396 |
+
"train_samples_per_second": 105.21,
|
| 397 |
+
"train_steps_per_second": 0.103
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 0,
|
| 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": 2.8922594581282816e+17,
|
| 418 |
+
"train_batch_size": 128,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/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/mc2_b32_224_mt5/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-32",
|
| 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/mc2/mc2_b32_224_mt5.pt",
|
| 56 |
+
"vocab_size": 151936
|
| 57 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.51.3"
|
| 14 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_224_mt5/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3be102397ff314138fd6db7bfc7d5ec1b82620410e32dbdcfc1070ed05f01ab
|
| 3 |
+
size 3744103520
|
continuous/qwen25/mc2_b32_224_mt5/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/mc2_b32_224_mt5/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/mc2_b32_224_mt5/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"extra_special_tokens": {},
|
| 203 |
+
"model_max_length": 8192,
|
| 204 |
+
"pad_token": "<|endoftext|>",
|
| 205 |
+
"padding_side": "right",
|
| 206 |
+
"split_special_tokens": false,
|
| 207 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 208 |
+
"unk_token": null
|
| 209 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/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": 18.851633071899414,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.8521,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 4.913137435913086,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.4668,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 3.711732864379883,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.1402,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.8085848093032837,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9606,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.3760971426963806,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.9259,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.3023693263530731,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.8769,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.2295050323009491,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.849,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.23235993087291718,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.8555,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.21183158457279205,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.8355,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.2282681167125702,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8154,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.20551615953445435,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8047,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.22275392711162567,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.816,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.20443852245807648,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8069,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.23271776735782623,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8081,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.22866295278072357,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.791,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.21614915132522583,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.7989,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.19388006627559662,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8004,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.2247370332479477,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8257,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.2218351811170578,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8176,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.24202752113342285,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8101,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.21109053492546082,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8018,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.2183699905872345,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.7893,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.21905075013637543,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.7915,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.224294975399971,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.785,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.22019147872924805,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8016,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.21728217601776123,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.20170770585536957,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.7888,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.2031199038028717,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.7887,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.21699602901935577,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8087,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.19951210916042328,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.7702,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.2116371989250183,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.779,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.21502335369586945,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.7842,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.21684308350086212,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.7814,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.21596458554267883,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.7949,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.22390969097614288,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.7932,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.21032102406024933,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.7807,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.2093779742717743,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8061,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.20874300599098206,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.7706,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.21508385241031647,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.7729,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.20019163191318512,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.7787,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.2038411647081375,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.7712,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.21196791529655457,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.7848,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.2145259529352188,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.7776,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.2022584229707718,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.7757,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.215616837143898,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7643,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.21615417301654816,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.7703,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.20064283907413483,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.7758,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.21319785714149475,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.7762,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.19214317202568054,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7621,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.1975613683462143,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.7721,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.21300330758094788,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7628,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.22670209407806396,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.7931,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.2229454219341278,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.7667,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.22615915536880493,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.7779,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.20635715126991272,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.7817,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.20528127253055573,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.7838,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.22709888219833374,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.7948,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.21890965104103088,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.7734,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.21296821534633636,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.7843,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.2007642686367035,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.7756,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.22354203462600708,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.7968,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.20954260230064392,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.7803,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.22417306900024414,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.7961,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.2038184255361557,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.7878,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.19295015931129456,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7456,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.23160135746002197,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.7788,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.22697637975215912,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.785,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.2047959566116333,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7544,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.20966047048568726,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.7763,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.21612700819969177,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.7832,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.21122533082962036,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7659,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.1929180771112442,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.7693,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.20580197870731354,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.7733,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.21322962641716003,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.7635,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.19792035222053528,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.7627,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.20933295786380768,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7552,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.2021162360906601,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.7723,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.1991892158985138,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.7688,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.2164875715970993,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.7559,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.20471495389938354,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.7833,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.21644623577594757,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.7749,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.20018397271633148,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.7698,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.20635634660720825,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7545,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.23111999034881592,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7606,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.2040008306503296,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.7717,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.20756450295448303,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7561,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.20789256691932678,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.7664,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.20008616149425507,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7616,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.20162472128868103,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.7611,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.19159908592700958,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.7642,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.22095756232738495,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.75,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.20940393209457397,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7527,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.22379320859909058,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7408,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.23747842013835907,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.7746,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.21515759825706482,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.788,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.21886976063251495,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.7746,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.2098701447248459,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7543,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.20617854595184326,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7605,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.21994440257549286,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7438,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.20533138513565063,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.7593,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.2080295830965042,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.7844,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.22447288036346436,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.7769,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.19973652064800262,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.7738,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.21467840671539307,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.7636,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.19389645755290985,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.7604,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.19176746904850006,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.7661,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.21307264268398285,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7566,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.20084352791309357,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.7804,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.2268911749124527,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.7641,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.21085506677627563,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.7782,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.21662864089012146,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.7738,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.22925730049610138,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.7698,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.2102561891078949,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7475,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.20776766538619995,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.7692,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.20996946096420288,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.7613,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.24222764372825623,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.7847,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.20822656154632568,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.7951,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.19593903422355652,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.7684,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.2149369865655899,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7482,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.21577879786491394,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7608,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.21942320466041565,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7628,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.21498684585094452,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.7796,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.20548290014266968,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.784,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.2004268765449524,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7521,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.20104455947875977,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7508,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.22390888631343842,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7598,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.245577871799469,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7426,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.2135520577430725,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7516,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.21334104239940643,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7616,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.20288154482841492,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.7644,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.20224247872829437,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.7754,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.21426266431808472,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7471,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.22922252118587494,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.755,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2138359099626541,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7619,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.20975469052791595,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7555,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.23630088567733765,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.773,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.21459327638149261,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.7701,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.2092815637588501,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.7772,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.22048142552375793,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7572,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.20790085196495056,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7578,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.21904468536376953,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.7707,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.22021040320396423,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7553,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.22167058289051056,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.7712,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.21744829416275024,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.7755,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.2162467986345291,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.7558,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.21672049164772034,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7624,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.20887450873851776,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7579,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.22213739156723022,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7537,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.20488682389259338,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7606,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.2022552341222763,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.7697,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.2104678750038147,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.7633,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.20754757523536682,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.7716,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.21007078886032104,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.7719,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.20623920857906342,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.777,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.2083931863307953,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.7635,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.2098359316587448,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.7792,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.21846245229244232,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.7756,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.22328051924705505,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.7653,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.18602155148983002,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.7672,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.19678039848804474,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7555,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.19770467281341553,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7381,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.2146969884634018,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7372,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.2084786742925644,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.7696,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.18722683191299438,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7577,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.2145388126373291,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.7684,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.21306298673152924,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7485,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.219868466258049,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.7692,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.21673241257667542,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.7646,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.2300117164850235,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7544,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.18959307670593262,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7533,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.21433012187480927,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.7665,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.21878719329833984,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7555,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.21097323298454285,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7612,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.21389278769493103,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7612,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.197654590010643,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7592,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.20099322497844696,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7475,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.2187747061252594,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.7906,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.22098729014396667,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.7755,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.20657196640968323,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.7867,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.21121002733707428,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7488,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.2191219925880432,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7425,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.19645315408706665,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.7643,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.20227861404418945,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.754,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.22106274962425232,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7614,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.2192857563495636,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.7709,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.20784002542495728,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.7657,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.19452053308486938,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7619,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.2010236531496048,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.7693,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.2189599871635437,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7517,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.2183946967124939,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.762,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.20032013952732086,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.7598,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.20691271126270294,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7549,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.21600258350372314,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.7696,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.22013059258460999,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.7762,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.21413011848926544,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.763,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.23020148277282715,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.781,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.21187059581279755,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.771,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.2177710086107254,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.7709,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.22378598153591156,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.7632,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.20693542063236237,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7553,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.21985894441604614,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7642,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.19816724956035614,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7513,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.20513850450515747,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7365,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.207875058054924,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7526,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.20086459815502167,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7574,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.1944563239812851,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7409,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.2288055121898651,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.7633,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.19181597232818604,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7483,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.19747528433799744,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.7685,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.20736488699913025,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7569,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.2138185203075409,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.7663,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.20999912917613983,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.7675,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.22371873259544373,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7531,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.21004252135753632,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.758,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.22466525435447693,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7491,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.20291705429553986,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.7672,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.2060854583978653,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.768,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.20523753762245178,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7586,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.19854310154914856,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.7665,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.215364471077919,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7614,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.21014845371246338,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7314,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.1947670429944992,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.7647,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.2089354544878006,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7603,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.19355863332748413,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.7709,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.20020000636577606,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.7561,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.210276260972023,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.745,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.208554208278656,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.753,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.21023084223270416,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7634,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.21077540516853333,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.7681,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.1951647400856018,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7463,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.1968516707420349,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7593,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.20563191175460815,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.7625,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.20538073778152466,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.7674,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.20538723468780518,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.7613,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.20573779940605164,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7458,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.20475490391254425,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7574,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.2250494360923767,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7502,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.2046728879213333,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7594,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.20852822065353394,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.7698,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.22431382536888123,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.7695,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.20519807934761047,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7499,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.21043139696121216,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.7753,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.2181980162858963,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.769,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.22241856157779694,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.7586,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.22627140581607819,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7564,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.20984125137329102,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.7762,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.22173526883125305,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.7711,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.20801933109760284,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7521,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.20693349838256836,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.7588,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.20497216284275055,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.756,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.20712344348430634,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7484,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.21236629784107208,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.7688,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.2061917781829834,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7546,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.19799219071865082,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7564,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.19740137457847595,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.7652,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.20053398609161377,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7574,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.20874102413654327,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7482,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.20321109890937805,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7396,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.21270829439163208,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.7713,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.522737300075184e+18,
|
| 1829 |
+
"train_loss": 0.7790380523073521,
|
| 1830 |
+
"train_runtime": 7608.5151,
|
| 1831 |
+
"train_samples_per_second": 87.389,
|
| 1832 |
+
"train_steps_per_second": 0.341
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 0,
|
| 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": 3.522737300075184e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b3386360cd284645e6e6ff2c530330e58c048c8b4d8c8a935807b6ccd7938ea
|
| 3 |
+
size 6136
|
continuous/qwen25/mc2_b32_224_mt5/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_224_mt5/pretrain/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"add_faster_video": false,
|
| 4 |
+
"add_time_instruction": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"bos_token_id": 151643,
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 1536,
|
| 15 |
+
"image_aspect_ratio": "square",
|
| 16 |
+
"image_crop_resolution": null,
|
| 17 |
+
"image_grid_pinpoints": null,
|
| 18 |
+
"image_split_resolution": null,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 8960,
|
| 21 |
+
"max_position_embeddings": 32768,
|
| 22 |
+
"max_window_layers": 21,
|
| 23 |
+
"mm_hidden_size": 768,
|
| 24 |
+
"mm_newline_position": "grid",
|
| 25 |
+
"mm_patch_merge_type": "flat",
|
| 26 |
+
"mm_projector_lr": null,
|
| 27 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 28 |
+
"mm_resampler_type": null,
|
| 29 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 30 |
+
"mm_spatial_pool_stride": null,
|
| 31 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 32 |
+
"mm_use_im_patch_token": false,
|
| 33 |
+
"mm_use_im_start_end": false,
|
| 34 |
+
"mm_vision_select_feature": "patch",
|
| 35 |
+
"mm_vision_select_layer": -2,
|
| 36 |
+
"mm_vision_tower": "open_clip_hub:ViT-B-32",
|
| 37 |
+
"mm_vision_tower_lr": null,
|
| 38 |
+
"model_type": "qwen2",
|
| 39 |
+
"num_attention_heads": 12,
|
| 40 |
+
"num_hidden_layers": 28,
|
| 41 |
+
"num_key_value_heads": 2,
|
| 42 |
+
"pos_skipping_range": 4096,
|
| 43 |
+
"rms_norm_eps": 1e-06,
|
| 44 |
+
"rope_scaling": null,
|
| 45 |
+
"rope_theta": 1000000.0,
|
| 46 |
+
"sliding_window": 32768,
|
| 47 |
+
"tie_word_embeddings": true,
|
| 48 |
+
"tokenizer_model_max_length": 8192,
|
| 49 |
+
"tokenizer_padding_side": "right",
|
| 50 |
+
"torch_dtype": "bfloat16",
|
| 51 |
+
"transformers_version": "4.51.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"use_mm_proj": true,
|
| 54 |
+
"use_pos_skipping": false,
|
| 55 |
+
"use_sliding_window": false,
|
| 56 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/mc2/mc2_b32_224_mt5.pt",
|
| 57 |
+
"vocab_size": 151936
|
| 58 |
+
}
|
continuous/qwen25/mc2_b32_224_mt5/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b16da43850476248c9557e9e5c6fafd8efef17a04c20e4337b8cf95d0fdd7a2
|
| 3 |
+
size 14170168
|
continuous/qwen25/mc2_b32_224_mt5/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": 16.188093185424805,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.4455,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.7068945169448853,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.4454,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.26814308762550354,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.2475,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.2884231209754944,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.1544,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.28855404257774353,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.0977,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.1947139948606491,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 3.0401,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.19985845685005188,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 2.9967,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.2204679250717163,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 2.9625,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.17692770063877106,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 2.9234,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.19658592343330383,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 2.9063,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.20924066007137299,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 2.8809,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.2425980418920517,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 2.8738,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.23350445926189423,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 2.8435,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.19085122644901276,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 2.833,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.19979968667030334,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 2.8315,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.1795835644006729,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 2.8125,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.24539010226726532,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.803,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.22495539486408234,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.783,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.22259913384914398,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.7664,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.2187742441892624,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.7371,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.2050088495016098,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.7248,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.20390349626541138,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.7287,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.20764009654521942,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.7162,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.22384804487228394,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.7043,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.20910479128360748,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.6965,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.1948527693748474,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.7064,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.1791394203901291,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.6821,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.19865986704826355,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.6872,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.21132099628448486,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.6725,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.20010480284690857,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.6713,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.19759146869182587,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.6808,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.20064064860343933,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.6742,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.21574664115905762,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.6601,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.1839200258255005,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.6649,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.20152194797992706,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.6642,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.23233142495155334,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.6661,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.1948752999305725,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.6612,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.19276322424411774,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.6666,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.201546773314476,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.6566,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.16680504381656647,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.6511,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.18057167530059814,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.6515,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.16592615842819214,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.6492,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.17419154942035675,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.6485,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.18064206838607788,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.6551,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.16944287717342377,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.6518,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.16752344369888306,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.6425,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.17169064283370972,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.6349,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.1774946004152298,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.638,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.1679491102695465,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.6476,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.17289473116397858,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.6521,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.17145980894565582,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.641,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.17962585389614105,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.6401,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.15820948779582977,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.6423,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.17105193436145782,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.6402,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.8922594581282816e+17,
|
| 394 |
+
"train_loss": 2.796132078958214,
|
| 395 |
+
"train_runtime": 5518.7955,
|
| 396 |
+
"train_samples_per_second": 101.132,
|
| 397 |
+
"train_steps_per_second": 0.099
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 0,
|
| 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": 2.8922594581282816e+17,
|
| 418 |
+
"train_batch_size": 128,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/mc2_b32_384/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/mc2_b32_384/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-32",
|
| 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/mc2/mc2_b32_384.pt",
|
| 56 |
+
"vocab_size": 151936
|
| 57 |
+
}
|
continuous/qwen25/mc2_b32_384/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.51.3"
|
| 14 |
+
}
|
continuous/qwen25/mc2_b32_384/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_384/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f1a0d339c2311ee5b4123625921d4c14d7973d140ee705d4caba24b8f934b2d
|
| 3 |
+
size 3744249448
|
continuous/qwen25/mc2_b32_384/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/mc2_b32_384/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/mc2_b32_384/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
+
"clean_up_tokenization_spaces": false,
|
| 200 |
+
"eos_token": "<|im_end|>",
|
| 201 |
+
"errors": "replace",
|
| 202 |
+
"extra_special_tokens": {},
|
| 203 |
+
"model_max_length": 8192,
|
| 204 |
+
"pad_token": "<|endoftext|>",
|
| 205 |
+
"padding_side": "right",
|
| 206 |
+
"split_special_tokens": false,
|
| 207 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 208 |
+
"unk_token": null
|
| 209 |
+
}
|
continuous/qwen25/mc2_b32_384/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": 15.383840560913086,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 1.8726,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 5.990599155426025,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.5492,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 3.4440715312957764,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.1526,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 0.6966694593429565,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 0.9539,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.33410629630088806,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 0.9137,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.2837993800640106,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.8655,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.23165304958820343,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.8378,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.23517096042633057,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.8452,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.21056026220321655,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.8259,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.22116126120090485,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.806,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.2048148512840271,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.7957,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.2243036925792694,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8072,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.20219188928604126,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.7965,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.2289455682039261,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.7992,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.23163682222366333,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.7802,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.21586012840270996,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.7898,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.19899813830852509,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.7911,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.22361953556537628,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8156,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.219087153673172,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8078,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.23868820071220398,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.7994,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.2075500786304474,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.7935,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.2141227275133133,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.7791,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.2181839644908905,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.782,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.2225862741470337,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.775,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.22540047764778137,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.7919,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.21362575888633728,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.7898,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.20368844270706177,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.7794,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.201960951089859,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.7774,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.21431373059749603,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.7988,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.1968095600605011,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.7601,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.20833563804626465,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.7687,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.22133943438529968,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.7743,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.21904130280017853,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.772,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.21186266839504242,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.785,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.21872499585151672,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.784,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.211822047829628,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.7712,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.20930512249469757,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.7969,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.20867615938186646,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.7623,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.2162156105041504,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.7634,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.1977560967206955,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.769,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.20526282489299774,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.7599,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.21280698478221893,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.7754,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.2148793637752533,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.7688,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.2062080353498459,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.7655,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.24711382389068604,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.7548,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.21217329800128937,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.761,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.19880926609039307,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.7663,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.21025267243385315,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.7655,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.19406861066818237,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.7518,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.20081454515457153,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.7638,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.21184001863002777,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.7532,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.2220362275838852,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.7833,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.21945007145404816,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.7576,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.22678475081920624,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.7686,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.21227458119392395,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.7724,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.20230905711650848,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.7743,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.22777555882930756,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.7837,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.22079899907112122,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.7629,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.20927117764949799,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.7738,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.20597979426383972,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.7656,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.22213640809059143,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.7859,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.21534180641174316,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.7705,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.21603569388389587,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.7861,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.20410752296447754,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.7788,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.19411912560462952,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7367,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.22498512268066406,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.768,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.2239718735218048,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.7767,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.20161114633083344,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7446,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.210236594080925,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.7662,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.21364259719848633,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.7733,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.20807692408561707,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.7564,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.1940847486257553,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.7609,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.20565570890903473,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.763,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.21103069186210632,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.753,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.19401995837688446,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.753,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.2197728306055069,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7453,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.20499007403850555,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.7627,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.19622552394866943,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.7595,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.218326136469841,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.7457,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.20219454169273376,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.773,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.2151292860507965,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.7645,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.19493094086647034,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.7595,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.20572160184383392,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7436,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.22591428458690643,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.7505,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.20739270746707916,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.763,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.2064555138349533,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7462,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.20591376721858978,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.7567,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.1989135444164276,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.7511,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.20371297001838684,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.7512,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.1890328973531723,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.7537,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.2217780202627182,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.7391,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.20734979212284088,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7434,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.21764840185642242,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.732,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.23452748358249664,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.7654,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.21479150652885437,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.7782,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.2257707118988037,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.7649,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.21748869121074677,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7436,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.2066296935081482,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.7509,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.21640074253082275,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7319,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.20519642531871796,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.7512,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.20733192563056946,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.7749,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.2192535698413849,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.7669,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.20466923713684082,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.7637,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.21867769956588745,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.7541,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.19291967153549194,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.751,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.19407761096954346,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.7558,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.21017566323280334,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.7458,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.19856701791286469,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.7699,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.22400452196598053,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.7549,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.21226583421230316,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.7689,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2126278132200241,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.7642,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.2233123481273651,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.7598,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.20903442800045013,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7372,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.20633214712142944,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.7599,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.20947207510471344,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.7504,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.23884429037570953,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.775,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.20783133804798126,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.7851,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.19370849430561066,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.7585,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.21001848578453064,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.738,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.2159092277288437,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.7509,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.21977759897708893,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.7536,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.2103213518857956,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.7691,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.20587781071662903,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.7737,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.204853817820549,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7422,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.19309115409851074,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.74,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2251758873462677,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7497,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.23439453542232513,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7314,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.20830710232257843,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7415,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.21194687485694885,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.7513,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.2015724629163742,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.7534,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.20338450372219086,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.7657,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.214559406042099,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7366,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.22325117886066437,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7448,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2132101058959961,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.7523,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.2216055691242218,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7462,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.23456403613090515,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.7626,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.21213336288928986,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.7584,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.19840936362743378,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.7666,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.21933071315288544,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7459,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.20734159648418427,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7466,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.21816565096378326,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.7608,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.2203170508146286,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7446,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.2188451737165451,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.7609,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.21702104806900024,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.7656,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.21808908879756927,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.7452,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.2162146121263504,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.7518,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.20347297191619873,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.7472,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.217868834733963,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.745,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.20065923035144806,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.7506,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.20200340449810028,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.7607,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.21017783880233765,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.7532,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.20562361180782318,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.7613,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.20941002666950226,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.7617,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.20227044820785522,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.7678,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.206011563539505,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.7531,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.20376558601856232,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.7678,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.2235708087682724,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.7653,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.22195963561534882,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.7547,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.18644508719444275,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.7575,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.19993478059768677,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.746,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.19814331829547882,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7276,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.21113994717597961,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.726,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.20446926355361938,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.7588,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.1857166737318039,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.7474,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.20969882607460022,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.7589,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.20913608372211456,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7385,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.21941456198692322,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.759,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.2136533111333847,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.7533,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.21895453333854675,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7444,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.189771369099617,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7433,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.21520698070526123,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.7566,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.2136349231004715,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7453,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.20940211415290833,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.7493,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.21608637273311615,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.7508,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.20055662095546722,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.7487,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.19987045228481293,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.737,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.21467405557632446,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.7811,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.21692128479480743,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.7655,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.2063300609588623,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.7756,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.20865792036056519,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7395,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.22162489593029022,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7317,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.19599907100200653,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.7532,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.19926801323890686,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7439,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.2178095132112503,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.7507,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.22496724128723145,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.761,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.20781144499778748,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.7559,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.19093890488147736,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.7502,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.20045556128025055,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.7593,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.22363367676734924,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7413,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.21335695683956146,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.7528,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.19817300140857697,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.7487,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.2070367932319641,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7445,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.21316714584827423,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.7578,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.21985910832881927,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.7653,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.21549537777900696,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.7535,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.22757412493228912,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.7709,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.20972946286201477,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.7612,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.21822333335876465,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.7603,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.2200002372264862,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.7531,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.20575149357318878,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7454,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.20693381130695343,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.7549,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.19586968421936035,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7415,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.20161943137645721,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.726,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.20567350089550018,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7433,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.20041579008102417,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7469,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.19666890799999237,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7309,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.2231721729040146,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.7531,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.19335560500621796,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7376,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.1987195611000061,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.7578,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.2108621597290039,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7467,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.21268270909786224,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.7573,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.2096831053495407,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.7574,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.22157461941242218,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7431,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.2138071060180664,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.7478,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.22422194480895996,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7385,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.20406831800937653,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.7568,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.2053954005241394,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.7564,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.2044302225112915,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7491,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.21215558052062988,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.7557,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.21803931891918182,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.7519,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.20937824249267578,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.721,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.19705185294151306,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.7548,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.21131522953510284,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.7506,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.1952614039182663,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.7611,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.1976848840713501,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.7449,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.20725642144680023,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7347,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.20657587051391602,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.742,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.20798185467720032,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.7539,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.20743981003761292,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.7565,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.1941622495651245,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7367,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.19321364164352417,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7482,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.20245927572250366,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.7519,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.20343516767024994,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.757,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.20187923312187195,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.749,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.20421932637691498,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7358,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.2028529793024063,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7466,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.22131693363189697,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7402,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.20323677361011505,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7484,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.20537400245666504,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.7613,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.22096484899520874,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.7605,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.20606841146945953,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7412,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.2065846472978592,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.765,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.21479660272598267,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.7598,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.2190456986427307,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.7486,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.2233254313468933,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7459,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.2083335667848587,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.7656,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.22133870422840118,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.7607,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.20623016357421875,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7421,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.2049606740474701,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.749,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.20565983653068542,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7464,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.2059585452079773,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7396,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.20890741050243378,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.7599,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.20203721523284912,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7446,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.19879069924354553,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7458,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.19786719977855682,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.7551,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.20486651360988617,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.747,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.20694822072982788,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7374,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.2013392299413681,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7285,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.211003378033638,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.7619,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.5229206575977595e+18,
|
| 1829 |
+
"train_loss": 0.7695491195505383,
|
| 1830 |
+
"train_runtime": 8030.1252,
|
| 1831 |
+
"train_samples_per_second": 82.801,
|
| 1832 |
+
"train_steps_per_second": 0.323
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 0,
|
| 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": 3.5229206575977595e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen25/mc2_b32_384/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:555ccf3aa534e7dd025b40a030ba92b17eea0c5cf8d928b8b714856b8c6e6f40
|
| 3 |
+
size 6136
|
continuous/qwen25/mc2_b32_384/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen25/mc2_b32_384/pretrain/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"add_faster_video": false,
|
| 4 |
+
"add_time_instruction": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Qwen2ForCausalLM"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"bos_token_id": 151643,
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 1536,
|
| 15 |
+
"image_aspect_ratio": "square",
|
| 16 |
+
"image_crop_resolution": null,
|
| 17 |
+
"image_grid_pinpoints": null,
|
| 18 |
+
"image_split_resolution": null,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 8960,
|
| 21 |
+
"max_position_embeddings": 32768,
|
| 22 |
+
"max_window_layers": 21,
|
| 23 |
+
"mm_hidden_size": 768,
|
| 24 |
+
"mm_newline_position": "grid",
|
| 25 |
+
"mm_patch_merge_type": "flat",
|
| 26 |
+
"mm_projector_lr": null,
|
| 27 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 28 |
+
"mm_resampler_type": null,
|
| 29 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 30 |
+
"mm_spatial_pool_stride": null,
|
| 31 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 32 |
+
"mm_use_im_patch_token": false,
|
| 33 |
+
"mm_use_im_start_end": false,
|
| 34 |
+
"mm_vision_select_feature": "patch",
|
| 35 |
+
"mm_vision_select_layer": -2,
|
| 36 |
+
"mm_vision_tower": "open_clip_hub:ViT-B-32",
|
| 37 |
+
"mm_vision_tower_lr": null,
|
| 38 |
+
"model_type": "qwen2",
|
| 39 |
+
"num_attention_heads": 12,
|
| 40 |
+
"num_hidden_layers": 28,
|
| 41 |
+
"num_key_value_heads": 2,
|
| 42 |
+
"pos_skipping_range": 4096,
|
| 43 |
+
"rms_norm_eps": 1e-06,
|
| 44 |
+
"rope_scaling": null,
|
| 45 |
+
"rope_theta": 1000000.0,
|
| 46 |
+
"sliding_window": 32768,
|
| 47 |
+
"tie_word_embeddings": true,
|
| 48 |
+
"tokenizer_model_max_length": 8192,
|
| 49 |
+
"tokenizer_padding_side": "right",
|
| 50 |
+
"torch_dtype": "bfloat16",
|
| 51 |
+
"transformers_version": "4.51.3",
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"use_mm_proj": true,
|
| 54 |
+
"use_pos_skipping": false,
|
| 55 |
+
"use_sliding_window": false,
|
| 56 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/mc2/mc2_b32_384.pt",
|
| 57 |
+
"vocab_size": 151936
|
| 58 |
+
}
|
continuous/qwen25/mc2_b32_384/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1734e6fd8833c109c9f3ffc87905e334e2a7e653805908951d03e1ea5d49856c
|
| 3 |
+
size 14170168
|
continuous/qwen25/mc2_b32_384/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": 4.913282871246338,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 4.3143,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.29135364294052124,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 3.3286,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.2493862807750702,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 3.1918,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.2661122977733612,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 3.1044,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.20425793528556824,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 3.0504,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.26044583320617676,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 2.9966,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.26813286542892456,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 2.9563,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.23848523199558258,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 2.9248,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.2570386528968811,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 2.8916,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.20815995335578918,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 2.8712,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.21847979724407196,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 2.8499,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.221582293510437,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 2.8405,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.2560037076473236,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 2.8115,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.21435058116912842,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 2.7901,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.24595656991004944,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 2.7717,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.22585666179656982,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 2.7352,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.22471407055854797,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 2.7147,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.23563247919082642,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 2.704,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.2158305048942566,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 2.6916,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.20536166429519653,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 2.6651,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.23894177377223969,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 2.6587,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.23598554730415344,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 2.6666,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.23176583647727966,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 2.6574,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.1723426878452301,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 2.6491,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.2095070481300354,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 2.6475,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.21966062486171722,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 2.6533,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.23358720541000366,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 2.629,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.21715745329856873,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 2.6351,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.18431982398033142,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 2.6229,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.17893564701080322,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 2.6253,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.18935273587703705,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 2.6293,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.22240467369556427,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 2.6244,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.2064782828092575,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 2.6105,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.25574684143066406,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 2.6156,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.19867505133152008,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 2.615,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.20089475810527802,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 2.6159,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.18496155738830566,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 2.6112,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.18805241584777832,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 2.6171,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.20461159944534302,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 2.6065,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.1977054476737976,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 2.5994,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.1823546439409256,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 2.6023,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.17597045004367828,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 2.5997,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.176083043217659,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 2.5958,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.21151046454906464,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 2.6086,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.17300966382026672,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 2.5981,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.1702403575181961,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 2.5896,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.18910983204841614,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 2.5842,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.1816013604402542,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 2.5872,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.19180288910865784,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 2.5963,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.1789848655462265,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 2.6018,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.18302041292190552,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 2.5917,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.17509768903255463,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 2.5908,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.16802935302257538,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 2.5909,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.17216719686985016,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 2.59,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.892409991363297e+17,
|
| 394 |
+
"train_loss": 2.7417463215119247,
|
| 395 |
+
"train_runtime": 5398.0675,
|
| 396 |
+
"train_samples_per_second": 103.394,
|
| 397 |
+
"train_steps_per_second": 0.101
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 0,
|
| 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": 2.892409991363297e+17,
|
| 418 |
+
"train_batch_size": 128,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen25/mc2_g14_224/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/mc2_g14_224/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": 1664,
|
| 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-bigG-14-worldwide",
|
| 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/mc2/mc2_g14_224.pt",
|
| 56 |
+
"vocab_size": 151936
|
| 57 |
+
}
|
continuous/qwen25/mc2_g14_224/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.51.3"
|
| 14 |
+
}
|
continuous/qwen25/mc2_g14_224/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|