Upload continuous evaluation checkpoints, batch 7
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +3 -0
- continuous/qwen3/dinov2_large/pretrain/mm_projector.bin +3 -0
- continuous/qwen3/dinov2_large/pretrain/trainer_state.json +421 -0
- continuous/qwen3/dinov2_small/finetune/added_tokens.json +28 -0
- continuous/qwen3/dinov2_small/finetune/chat_template.jinja +89 -0
- continuous/qwen3/dinov2_small/finetune/config.json +89 -0
- continuous/qwen3/dinov2_small/finetune/generation_config.json +12 -0
- continuous/qwen3/dinov2_small/finetune/merges.txt +0 -0
- continuous/qwen3/dinov2_small/finetune/model.safetensors +3 -0
- continuous/qwen3/dinov2_small/finetune/special_tokens_map.json +31 -0
- continuous/qwen3/dinov2_small/finetune/tokenizer.json +3 -0
- continuous/qwen3/dinov2_small/finetune/tokenizer_config.json +240 -0
- continuous/qwen3/dinov2_small/finetune/trainer_state.json +1856 -0
- continuous/qwen3/dinov2_small/finetune/training_args.bin +3 -0
- continuous/qwen3/dinov2_small/finetune/vocab.json +0 -0
- continuous/qwen3/dinov2_small/pretrain/config.json +89 -0
- continuous/qwen3/dinov2_small/pretrain/mm_projector.bin +3 -0
- continuous/qwen3/dinov2_small/pretrain/trainer_state.json +421 -0
- continuous/qwen3/dinov3_vitl16/finetune/added_tokens.json +28 -0
- continuous/qwen3/dinov3_vitl16/finetune/chat_template.jinja +89 -0
- continuous/qwen3/dinov3_vitl16/finetune/config.json +89 -0
- continuous/qwen3/dinov3_vitl16/finetune/generation_config.json +12 -0
- continuous/qwen3/dinov3_vitl16/finetune/merges.txt +0 -0
- continuous/qwen3/dinov3_vitl16/finetune/model.safetensors +3 -0
- continuous/qwen3/dinov3_vitl16/finetune/special_tokens_map.json +31 -0
- continuous/qwen3/dinov3_vitl16/finetune/tokenizer.json +3 -0
- continuous/qwen3/dinov3_vitl16/finetune/tokenizer_config.json +240 -0
- continuous/qwen3/dinov3_vitl16/finetune/trainer_state.json +1856 -0
- continuous/qwen3/dinov3_vitl16/finetune/training_args.bin +3 -0
- continuous/qwen3/dinov3_vitl16/finetune/vocab.json +0 -0
- continuous/qwen3/dinov3_vitl16/pretrain/config.json +89 -0
- continuous/qwen3/dinov3_vitl16/pretrain/mm_projector.bin +3 -0
- continuous/qwen3/dinov3_vitl16/pretrain/trainer_state.json +421 -0
- continuous/qwen3/eupe_convnext_b/finetune/added_tokens.json +28 -0
- continuous/qwen3/eupe_convnext_b/finetune/chat_template.jinja +89 -0
- continuous/qwen3/eupe_convnext_b/finetune/config.json +89 -0
- continuous/qwen3/eupe_convnext_b/finetune/generation_config.json +12 -0
- continuous/qwen3/eupe_convnext_b/finetune/merges.txt +0 -0
- continuous/qwen3/eupe_convnext_b/finetune/model.safetensors +3 -0
- continuous/qwen3/eupe_convnext_b/finetune/special_tokens_map.json +31 -0
- continuous/qwen3/eupe_convnext_b/finetune/tokenizer.json +3 -0
- continuous/qwen3/eupe_convnext_b/finetune/tokenizer_config.json +240 -0
- continuous/qwen3/eupe_convnext_b/finetune/trainer_state.json +1856 -0
- continuous/qwen3/eupe_convnext_b/finetune/training_args.bin +3 -0
- continuous/qwen3/eupe_convnext_b/finetune/vocab.json +0 -0
- continuous/qwen3/eupe_convnext_b/pretrain/config.json +89 -0
- continuous/qwen3/eupe_convnext_b/pretrain/mm_projector.bin +3 -0
- continuous/qwen3/eupe_convnext_b/pretrain/trainer_state.json +421 -0
- continuous/qwen3/eupe_vit_b/finetune/added_tokens.json +28 -0
- continuous/qwen3/eupe_vit_b/finetune/chat_template.jinja +89 -0
.gitattributes
CHANGED
|
@@ -119,3 +119,6 @@ continuous/qwen3/dino_vits8/finetune/tokenizer.json filter=lfs diff=lfs merge=lf
|
|
| 119 |
continuous/qwen3/dinov2_base/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 120 |
continuous/qwen3/dinov2_giant/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 121 |
continuous/qwen3/dinov2_large/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 119 |
continuous/qwen3/dinov2_base/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 120 |
continuous/qwen3/dinov2_giant/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 121 |
continuous/qwen3/dinov2_large/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 122 |
+
continuous/qwen3/dinov2_small/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 123 |
+
continuous/qwen3/dinov3_vitl16/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 124 |
+
continuous/qwen3/eupe_convnext_b/finetune/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
continuous/qwen3/dinov2_large/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58a1e5337b13ab56e5feeeea74e0437a9e547c9818eb7d440aaa4cc83ec95409
|
| 3 |
+
size 25184312
|
continuous/qwen3/dinov2_large/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": 3.935349464416504,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 9.2703,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.3469666838645935,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 6.0963,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.10113006830215454,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 5.0508,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.06623757630586624,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 4.7965,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.04242463782429695,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 4.7212,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.0354999415576458,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 4.6771,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.03064095228910446,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 4.6335,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.05373946204781532,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 4.6124,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.04035679250955582,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 4.5763,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.06383256614208221,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 4.5531,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.058029983192682266,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 4.5267,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.0790121853351593,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 4.5129,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.10581627488136292,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 4.4875,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.05516836792230606,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 4.4672,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.08047036826610565,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 4.4609,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.07355165481567383,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 4.4438,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.06547911465167999,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 4.4267,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.09399333596229553,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 4.4125,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.08643694221973419,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 4.4095,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.09929905831813812,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 4.3882,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.09076589345932007,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 4.3797,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.0823846384882927,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 4.3902,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.07077837735414505,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 4.3893,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.07659544050693512,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 4.3732,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.07143983989953995,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 4.3543,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.0845206156373024,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 4.3642,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.08939521014690399,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 4.346,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.06501373648643494,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 4.3439,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.07081393897533417,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 4.3279,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.10623709112405777,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 4.3299,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.06832797080278397,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 4.3455,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.07298921793699265,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 4.3342,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.0745057463645935,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 4.3162,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.053755417466163635,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 4.3165,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.05914594233036041,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 4.3247,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.062051113694906235,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 4.331,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.05780702829360962,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 4.314,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.07895472645759583,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 4.3308,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.11759057641029358,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 4.3118,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.066788449883461,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 4.3158,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.056662146002054214,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 4.3162,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.062007199972867966,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 4.3052,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.05091783031821251,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 4.3125,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.07110366970300674,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 4.3184,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.060737378895282745,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 4.3112,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.06588970124721527,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 4.3021,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.05576775595545769,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 4.2928,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.04806851968169212,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 4.3021,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.050398342311382294,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 4.3008,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.04615636542439461,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 4.3147,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.04976646602153778,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 4.3014,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.04700024425983429,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 4.2935,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.04648818075656891,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 4.2982,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.05095994099974632,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 4.304,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 3.291865432624988e+17,
|
| 394 |
+
"train_loss": 4.528371506437249,
|
| 395 |
+
"train_runtime": 5029.0761,
|
| 396 |
+
"train_samples_per_second": 110.98,
|
| 397 |
+
"train_steps_per_second": 0.108
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 31896544,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 3.291865432624988e+17,
|
| 418 |
+
"train_batch_size": 32,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen3/dinov2_small/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
continuous/qwen3/dinov2_small/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|
continuous/qwen3/dinov2_small/finetune/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 384,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": 0.0001,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/dinov2/dinov2-small",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": null,
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/dinov2_small/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"temperature": 0.6,
|
| 9 |
+
"top_k": 20,
|
| 10 |
+
"top_p": 0.95,
|
| 11 |
+
"transformers_version": "4.57.6"
|
| 12 |
+
}
|
continuous/qwen3/dinov2_small/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/dinov2_small/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1c958708d56bdc6a690dbfcf140cbb69c8951c4320a5606410a6d82ee639f7c
|
| 3 |
+
size 4127600424
|
continuous/qwen3/dinov2_small/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/qwen3/dinov2_small/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
continuous/qwen3/dinov2_small/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"clean_up_tokenization_spaces": false,
|
| 231 |
+
"eos_token": "<|im_end|>",
|
| 232 |
+
"errors": "replace",
|
| 233 |
+
"extra_special_tokens": {},
|
| 234 |
+
"model_max_length": 8192,
|
| 235 |
+
"pad_token": "<|endoftext|>",
|
| 236 |
+
"padding_side": "right",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
continuous/qwen3/dinov2_small/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": 15.624537467956543,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 2.2342,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 6.836838722229004,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.8927,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 4.381350994110107,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.5238,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 1.1571054458618164,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 1.1955,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.8091592788696289,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 1.0918,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.4348328411579132,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 1.0017,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.31968003511428833,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.9519,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.3127453923225403,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.9482,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.32128918170928955,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.9157,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.25975027680397034,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8961,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.3117559254169464,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8769,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.2634032368659973,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8862,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.2560584843158722,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8821,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.28594592213630676,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8745,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.32725322246551514,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.8605,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.26077333092689514,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8651,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.23693934082984924,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8649,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.2654498815536499,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8934,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.256147176027298,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.882,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.2876957058906555,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8714,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.2683227062225342,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8652,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.2618764042854309,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8543,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.26411810517311096,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.8542,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.2556880712509155,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.8488,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.2852674424648285,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8655,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.2530583143234253,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.862,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.24020376801490784,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.8492,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.25140827894210815,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8499,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.2532294690608978,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8665,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.22451134026050568,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.8265,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.24285843968391418,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.8411,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.24850668013095856,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.8404,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.2538253366947174,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.8398,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.24655264616012573,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.851,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.25984036922454834,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.8513,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.24669358134269714,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.8373,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.26052263379096985,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8637,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.24186530709266663,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.8233,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.24759501218795776,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.8321,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.2396448403596878,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.8356,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.22981920838356018,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.8238,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.24522818624973297,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.8394,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.25462809205055237,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.8289,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.2471320927143097,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.8271,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.2506164014339447,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.8166,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.23979805409908295,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.8216,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.24916283786296844,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.8308,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.2389727234840393,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.8293,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.24163521826267242,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.8154,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.22304855287075043,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.8198,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.23337753117084503,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.8161,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.24810390174388885,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.8458,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.2534521222114563,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.8165,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.2516704201698303,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.8304,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.22738400101661682,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.8321,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.22211839258670807,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.8371,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.265154093503952,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8463,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.2583479881286621,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.826,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.253694623708725,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.8366,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.23365136981010437,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.8273,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.2572700083255768,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.8473,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.24246062338352203,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.8313,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.24782989919185638,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8478,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.2364404797554016,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.8359,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.22487109899520874,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7931,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.2581082880496979,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.8316,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.29244348406791687,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.8354,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.24163450300693512,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.8057,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.2297225147485733,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.8312,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.23769329488277435,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.8356,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.23697109520435333,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.8148,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.22743827104568481,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.8192,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.24342577159404755,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.8254,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.23213458061218262,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.8143,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.21926604211330414,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.8126,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.24811910092830658,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.806,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.23229680955410004,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.8236,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.23288407921791077,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.82,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.24593591690063477,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.805,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.22806844115257263,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.8323,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.24673622846603394,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.8251,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.24198167026042938,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.8177,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.25008395314216614,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.804,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.24805831909179688,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.8101,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.23838824033737183,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.8196,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.22993740439414978,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.8037,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.2425903081893921,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.8144,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.22697564959526062,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.8083,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.22871054708957672,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.808,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.22377772629261017,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.8155,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.24663494527339935,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.8051,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.23671011626720428,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.802,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.26396051049232483,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7883,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.25190505385398865,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.8225,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.2467850297689438,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8374,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.25506702065467834,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.8209,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.24526450037956238,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.8023,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.23464852571487427,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.8068,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.24718523025512695,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.795,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.23248830437660217,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.8093,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.2375115305185318,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.8327,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.2531421184539795,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.8248,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.24159301817417145,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.8218,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.25029894709587097,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.8126,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.214462012052536,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.8087,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.23142372071743011,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.8159,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.23929022252559662,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.8074,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.22869950532913208,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.8315,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.26720696687698364,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.811,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.2499423325061798,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.826,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2454608827829361,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.8221,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.25375068187713623,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.8176,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.2418401539325714,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7971,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.23295573890209198,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.8187,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.225255087018013,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.8088,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.2688584327697754,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.8315,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.22499005496501923,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8431,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.22472895681858063,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.8179,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.24885492026805878,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7985,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.24628250300884247,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.8072,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.25121641159057617,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.8112,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.24759140610694885,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.8282,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.23695331811904907,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.8291,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.23264068365097046,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.801,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.22890765964984894,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7992,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2566912770271301,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.8045,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.26178210973739624,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7926,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.24633564054965973,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7983,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.2392113357782364,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.8097,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.2307208776473999,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.8096,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.23572251200675964,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.8248,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.2446363866329193,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7928,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.25272423028945923,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.8045,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2477063685655594,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.8098,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.2449280321598053,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.8005,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.27120018005371094,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.8205,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.2613038718700409,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.8202,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.2361697405576706,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.824,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.25068172812461853,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.8047,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.23269347846508026,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.803,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.24546365439891815,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.8183,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.2524162530899048,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.8053,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.26701581478118896,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.8189,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.2653197646141052,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.8202,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.26476120948791504,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.8079,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.2423555850982666,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.8092,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.2341759353876114,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.8079,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.25143906474113464,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.8024,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.22958077490329742,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.8093,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.23456212878227234,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.8154,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.23632965981960297,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.8096,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.23934778571128845,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.818,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.23708145320415497,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.8212,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.22899021208286285,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.8232,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.23866713047027588,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.8098,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.23293018341064453,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.8299,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.2464074045419693,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.8228,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.25619637966156006,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.813,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.21232886612415314,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.815,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.2248295396566391,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.8018,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.21958424150943756,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7848,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.2375350147485733,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7841,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.24179482460021973,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.8141,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.21355244517326355,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.8047,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.24236154556274414,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.8131,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.23536023497581482,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7929,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.2514728307723999,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.8147,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.24019326269626617,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.8135,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.2569262385368347,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.8022,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.21748505532741547,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.8012,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.25000280141830444,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.8149,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.25160646438598633,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.8051,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.2377699911594391,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.8128,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.2335338145494461,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.8095,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.24054065346717834,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.805,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.22750678658485413,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7939,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.25513651967048645,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8361,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.23707197606563568,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.8229,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.2320546954870224,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.8337,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.24712525308132172,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.7978,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.24343638122081757,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7887,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.2250574678182602,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.8146,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.23629406094551086,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.8015,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.24868246912956238,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.8056,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.2521745264530182,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.8193,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.2320464849472046,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.8107,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.24508513510227203,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.8085,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.22727929055690765,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.8149,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.24834538996219635,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7991,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.2402934581041336,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.8059,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.2339668720960617,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.8054,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.2417771816253662,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7972,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.24816003441810608,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.8189,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.23272517323493958,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.8247,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.2444257289171219,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.8103,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.25344401597976685,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.8292,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.24064579606056213,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.8165,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.2536354064941406,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.8168,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.25352364778518677,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.8088,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.2339554876089096,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.801,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.22402164340019226,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.8082,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.2286897450685501,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7972,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.23173454403877258,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7847,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.23831601440906525,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.8007,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.22704723477363586,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.8044,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.21436303853988647,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7922,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.259844034910202,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.814,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.23456193506717682,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7925,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.23511451482772827,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.8142,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.23682047426700592,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.8032,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.23580284416675568,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.81,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.2417943775653839,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.8128,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.25092706084251404,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.8023,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.24193722009658813,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.8047,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.24771784245967865,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7981,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.22948430478572845,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.8136,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.22554738819599152,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.8161,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.24178612232208252,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.8049,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.24119937419891357,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.8159,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.250855416059494,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.8064,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.2328377366065979,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.779,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.22472797334194183,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.8137,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.24901777505874634,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.8058,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.23691897094249725,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.8171,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.23950277268886566,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.8023,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.24720610678195953,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7915,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.2436911165714264,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.8037,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.22815220057964325,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.8094,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.2316659837961197,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.8176,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.22374197840690613,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.793,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.223164901137352,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.8038,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.22610574960708618,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.8118,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.23110342025756836,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.8159,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.2406763732433319,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.8114,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.23341716825962067,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7902,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.22817808389663696,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.8046,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.2619462311267853,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7965,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.23686696588993073,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.8046,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.23332513868808746,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.8129,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.2537088096141815,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.815,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.2419670820236206,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7984,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.22614826261997223,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.8209,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.23442107439041138,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.8166,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.24654094874858856,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.8065,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.25365033745765686,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.8041,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.2372145652770996,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.8234,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.24561412632465363,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.8187,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.26081693172454834,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.8011,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.2921110689640045,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.8066,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.23282940685749054,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7991,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.23163165152072906,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7925,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.23379896581172943,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.814,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.23051710426807404,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7994,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.22218111157417297,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.8028,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.22532738745212555,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.8127,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.2335330694913864,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.8051,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.23367281258106232,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7929,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.21525059640407562,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7879,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.24751196801662445,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.817,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.6099922794077225e+18,
|
| 1829 |
+
"train_loss": 0.834926181818184,
|
| 1830 |
+
"train_runtime": 12901.749,
|
| 1831 |
+
"train_samples_per_second": 51.536,
|
| 1832 |
+
"train_steps_per_second": 0.201
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 418855352,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 3.6099922794077225e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen3/dinov2_small/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:720e07430d6282742e7919837d8d0f846cbe1c0ad7fffe2e73b97f525f8922c5
|
| 3 |
+
size 6264
|
continuous/qwen3/dinov2_small/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/dinov2_small/pretrain/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 384,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": null,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "hf:/cache/ckpt/download/tokenizer/continuous/dinov2/dinov2-small",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": null,
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/dinov2_small/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0572ebdba966b24bcd7c5268e68afa50e70b26aa779d01f6e155f770c6a0c18c
|
| 3 |
+
size 19941432
|
continuous/qwen3/dinov2_small/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": 1.981568455696106,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 10.0091,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 1.2175110578536987,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 7.2139,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.27913111448287964,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 5.8258,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.12027377635240555,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 5.1439,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.057079631835222244,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 4.9147,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.04072239249944687,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 4.8195,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.03329303860664368,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 4.7548,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.039421044290065765,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 4.7156,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.04835725575685501,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 4.6777,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.05158976837992668,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 4.6521,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.05765845626592636,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 4.6208,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.05552615225315094,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 4.6098,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.055311545729637146,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 4.5768,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.049433622509241104,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 4.5581,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.0648411363363266,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 4.5522,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.056050561368465424,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 4.5355,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.059597618877887726,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 4.517,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.06521910429000854,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 4.503,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.09421490877866745,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 4.4987,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.06330699473619461,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 4.4762,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.051724717020988464,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 4.4674,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.08207134902477264,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 4.4808,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.06207136809825897,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 4.4778,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.04402885586023331,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 4.4623,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.06342238187789917,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 4.4449,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.051453039050102234,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 4.4551,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.07701152563095093,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 4.4342,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.06837359815835953,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 4.4318,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.05755666270852089,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 4.4146,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.06796176731586456,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 4.4144,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.06201424449682236,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 4.4298,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.05542974919080734,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 4.4216,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.06300622969865799,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 4.4029,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.057855937629938126,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 4.4001,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.05278908461332321,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 4.4115,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.06046174466609955,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 4.4158,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.053728338330984116,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 4.3976,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.04710477218031883,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 4.4145,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.04968581721186638,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 4.3998,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.0495472326874733,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 4.3986,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.04661352559924126,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 4.4005,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.054816871881484985,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 4.3905,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.05378881096839905,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 4.397,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.05791974067687988,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 4.4033,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.053681630641222,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 4.3943,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.05295180529356003,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 4.3893,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.04712679237127304,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 4.3744,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.049998342990875244,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 4.3848,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.05192194879055023,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 4.3859,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.04595423862338066,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 4.3986,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.05314210057258606,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 4.3863,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.04340951889753342,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 4.3789,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.045608021318912506,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 4.3817,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.04761801287531853,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 4.3877,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.7490702199042867e+17,
|
| 394 |
+
"train_loss": 4.668020983354761,
|
| 395 |
+
"train_runtime": 5106.7545,
|
| 396 |
+
"train_samples_per_second": 109.292,
|
| 397 |
+
"train_steps_per_second": 0.107
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 31896544,
|
| 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": 2.7490702199042867e+17,
|
| 418 |
+
"train_batch_size": 32,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen3/dinov3_vitl16/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
continuous/qwen3/dinov3_vitl16/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|
continuous/qwen3/dinov3_vitl16/finetune/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 1024,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": 0.0001,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "vtb_ssl:dinov3",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/dinov3/dinov3_vitl16_pretrain_lvd1689m-8aa4cbdd.pth",
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/dinov3_vitl16/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"temperature": 0.6,
|
| 9 |
+
"top_k": 20,
|
| 10 |
+
"top_p": 0.95,
|
| 11 |
+
"transformers_version": "4.57.6"
|
| 12 |
+
}
|
continuous/qwen3/dinov3_vitl16/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/dinov3_vitl16/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d49efdcb6bfc423d75bb0f53685cb68b60a33e9389603e4150db4619e512b772
|
| 3 |
+
size 4695201696
|
continuous/qwen3/dinov3_vitl16/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/qwen3/dinov3_vitl16/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
continuous/qwen3/dinov3_vitl16/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"clean_up_tokenization_spaces": false,
|
| 231 |
+
"eos_token": "<|im_end|>",
|
| 232 |
+
"errors": "replace",
|
| 233 |
+
"extra_special_tokens": {},
|
| 234 |
+
"model_max_length": 8192,
|
| 235 |
+
"pad_token": "<|endoftext|>",
|
| 236 |
+
"padding_side": "right",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
continuous/qwen3/dinov3_vitl16/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": 23.31719398498535,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 2.4028,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 7.90120267868042,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 1.8772,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 4.564217567443848,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.5052,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 1.6896857023239136,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 1.1912,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.8898191452026367,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 1.0776,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.4539790153503418,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.9906,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.3819841146469116,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.9421,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.3581533432006836,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.9393,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.3372410833835602,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.9081,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.3283670246601105,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8895,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.3054846227169037,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8697,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.29848620295524597,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8799,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.28082573413848877,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8773,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.33736637234687805,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8698,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.33269011974334717,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.8556,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.3210979104042053,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8611,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.26441463828086853,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8609,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.3101763427257538,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8891,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.29200074076652527,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8762,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.33898645639419556,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8671,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.2869936525821686,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8611,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.29862067103385925,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8498,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.3657817244529724,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.8505,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.28119269013404846,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.8449,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.33612650632858276,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8613,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.27853769063949585,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.858,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.28210505843162537,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.846,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.27124395966529846,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8458,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.29515722393989563,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8628,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.25983545184135437,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.8234,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.27590227127075195,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.8377,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.27851027250289917,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.8367,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.2809886634349823,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.836,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.2639006972312927,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.847,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.29636120796203613,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.8465,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.27102580666542053,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.8338,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.28004640340805054,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8606,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.27039802074432373,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.8198,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.2747540771961212,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.8288,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.2734030783176422,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.8312,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.28341948986053467,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.8204,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.28048306703567505,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.8358,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.27600643038749695,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.8256,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.27038583159446716,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.8234,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.28336140513420105,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.813,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.2814617156982422,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.818,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.2727006673812866,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.8264,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.26246950030326843,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.8255,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.2608005404472351,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.8118,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.25541257858276367,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.817,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.274120569229126,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.8126,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.2798087000846863,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.8416,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.286236971616745,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.8139,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.2856202721595764,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.8272,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.2475760281085968,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.8288,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.24235163629055023,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.8338,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.3027414381504059,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8425,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.2829147279262543,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.8218,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.2718786597251892,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.8326,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.26542165875434875,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.8232,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.27543461322784424,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.8433,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.27215585112571716,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.8272,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.26917165517807007,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8438,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.2493184506893158,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.8323,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.25198912620544434,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7896,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.27944087982177734,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.827,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.3353561758995056,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.8321,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.25417837500572205,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.8022,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.26473468542099,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.8272,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.26961833238601685,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.832,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.2553854286670685,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.8115,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.2429138422012329,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.8156,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.2645387649536133,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.8223,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.2643314003944397,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.81,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.24028368294239044,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.8094,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.2808202803134918,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.8022,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.278594434261322,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.8205,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.2576900124549866,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.8165,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.2727756202220917,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.8016,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.2712702751159668,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.8286,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.2652953267097473,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.8212,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.2616623640060425,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.8136,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.25866496562957764,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7998,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.2767125070095062,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.8058,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.26357096433639526,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.8161,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.262279748916626,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.8004,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.2606140375137329,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.8105,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.2364703267812729,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.8052,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.24938265979290009,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.8049,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.2361527532339096,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.8114,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.2797771692276001,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.8013,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.2574062943458557,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7984,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.2764701247215271,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7846,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.28892436623573303,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.8185,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.2610552906990051,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8335,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.2642318308353424,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.817,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.2601834833621979,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7977,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.24889472126960754,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.8032,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.27910688519477844,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7907,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.2517434060573578,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.8059,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.26551511883735657,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.8291,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.2739312946796417,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.8206,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.260257363319397,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.8179,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.2643580436706543,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.8082,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.2297937273979187,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.8042,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.249811589717865,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.8124,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.2596070170402527,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.8043,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.29561904072761536,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.8275,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.2931879758834839,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.8066,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.2762984335422516,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.8222,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2672664523124695,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.8179,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.28746068477630615,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.8135,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.26635175943374634,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.7925,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.25456395745277405,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.8139,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.2387772798538208,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.8044,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.29458409547805786,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.8271,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.2532731890678406,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8388,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.24544808268547058,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.8135,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.27274778485298157,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7933,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.25690412521362305,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.8027,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.26950547099113464,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.8074,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.2699414789676666,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.8239,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.25470924377441406,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.825,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.2527425289154053,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.796,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.24426846206188202,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7955,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2792189419269562,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7989,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.30487290024757385,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7885,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.28188785910606384,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7942,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.2661869525909424,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.8057,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.24017399549484253,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.8055,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.2512076497077942,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.8203,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.2705794870853424,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7885,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.27781105041503906,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7996,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.2715962529182434,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.805,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.2520642578601837,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7963,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.2866469919681549,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.8162,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.2779085040092468,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.816,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.24743129312992096,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.8202,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.2693771719932556,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7999,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.2603982985019684,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7988,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.27160248160362244,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.8133,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.2742234766483307,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.8002,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.2918796241283417,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.8146,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.2763383090496063,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.8166,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.29815933108329773,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.8034,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.2554953992366791,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.8045,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.26394233107566833,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.8033,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.29195624589920044,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7986,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.25344163179397583,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.8047,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.24963697791099548,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.8111,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.252204030752182,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.8059,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.25829601287841797,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.8136,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.25951188802719116,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.8167,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.25496795773506165,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.8189,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.25484028458595276,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.8064,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.261547327041626,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.8252,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.27315887808799744,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.8178,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.26028379797935486,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.8082,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.22904863953590393,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.8098,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.24978993833065033,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7972,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.24164029955863953,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7801,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.2569252848625183,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.78,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.2556617259979248,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.8096,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.23273445665836334,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.8001,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.25433436036109924,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.8084,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.2612954080104828,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7886,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.26980724930763245,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.8109,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.26813068985939026,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.8092,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.2719709575176239,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7976,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.2305494248867035,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.7967,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.27264702320098877,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.8106,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.27427124977111816,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.8002,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.26595818996429443,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.8078,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.2548576593399048,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.8044,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.2622431814670563,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.8008,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.24583229422569275,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7891,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.26947030425071716,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8314,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.2490374892950058,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.8187,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.24919630587100983,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.829,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.2676793932914734,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.793,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.25945839285850525,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7843,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.24651575088500977,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.809,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.24769631028175354,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7967,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.27481865882873535,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.8013,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.272874116897583,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.8142,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.2592027485370636,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.8061,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.26810386776924133,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.8039,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.25228944420814514,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.8099,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.26220640540122986,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7947,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.25706183910369873,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.801,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.24810799956321716,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.8001,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.25677919387817383,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7922,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.275153249502182,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.814,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.2524268925189972,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.8198,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.2608914077281952,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.8057,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.2757207155227661,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.8246,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.25463932752609253,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.8122,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.26352038979530334,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.8119,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.27641060948371887,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.8047,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.24887239933013916,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7964,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.2444104701280594,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.8034,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.23743781447410583,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.7924,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.24710287153720856,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.7803,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.2598738968372345,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7955,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.24742946028709412,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7996,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.23936721682548523,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7876,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.28171291947364807,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.8093,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.2620701491832733,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7872,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.25030815601348877,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.8096,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.2571042776107788,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7987,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.2515510618686676,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.8062,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.2598980665206909,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.8087,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.2800433039665222,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7979,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.2612910866737366,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.8003,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.2640438973903656,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.7935,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.2590593099594116,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.8095,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.24720467627048492,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.8114,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.2677866220474243,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.8008,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.25306442379951477,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.8104,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.2846547067165375,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.8018,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.2567547857761383,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7746,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.24811720848083496,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.8087,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.2700637876987457,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.8006,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.24908271431922913,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.8124,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.2570423185825348,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.7968,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.25494062900543213,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7863,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.26757436990737915,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7987,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.25992438197135925,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.8057,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.24861054122447968,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.8127,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.2340226173400879,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7877,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.23632735013961792,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.8001,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.24868935346603394,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.8076,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.2610982656478882,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.8111,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.25873619318008423,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.8067,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.2583012878894806,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7856,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.24566426873207092,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.8,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.28499770164489746,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7912,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.2527550756931305,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.8,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.24570052325725555,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.8082,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.2687113881111145,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.8106,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.25470414757728577,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7935,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.2553342580795288,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.8157,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.2556399703025818,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.8119,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.2798268795013428,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.8018,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.27164730429649353,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7995,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.270422101020813,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.8186,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.2712685763835907,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.8141,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.2596576511859894,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.7963,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.3044552803039551,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.8021,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.24800874292850494,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7943,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.26155510544776917,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7882,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.261965274810791,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.8099,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.24861469864845276,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7944,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.24160204827785492,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7982,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.24786590039730072,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.808,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.2638911306858063,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.8002,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.2500248849391937,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7884,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.24325521290302277,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7841,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.26864781975746155,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.8123,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 4.319721686829105e+18,
|
| 1829 |
+
"train_loss": 0.8310841814481069,
|
| 1830 |
+
"train_runtime": 13086.7791,
|
| 1831 |
+
"train_samples_per_second": 50.807,
|
| 1832 |
+
"train_steps_per_second": 0.198
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 418855352,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 4.319721686829105e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen3/dinov3_vitl16/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c82f5ecada46bd19f1287b8f5002b807813bef62ff81a8379cdc4fe5cc39f9c2
|
| 3 |
+
size 6264
|
continuous/qwen3/dinov3_vitl16/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/dinov3_vitl16/pretrain/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 1024,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": null,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "vtb_ssl:dinov3",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/dinov3/dinov3_vitl16_pretrain_lvd1689m-8aa4cbdd.pth",
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/dinov3_vitl16/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6b7351dedc110276b5b9700afc686a4c940c9fa7e043fed7b81dc82290d4354
|
| 3 |
+
size 25184312
|
continuous/qwen3/dinov3_vitl16/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": 4.92300271987915,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 11.1801,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 0.479547917842865,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 7.8922,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.4569433629512787,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 5.9465,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.15538771450519562,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 5.1033,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.06296360492706299,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 4.87,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.045744821429252625,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 4.7992,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.04112619161605835,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 4.7414,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.057952143251895905,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 4.7134,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.056048035621643066,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 4.6786,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.06540649384260178,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 4.6609,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.056451477110385895,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 4.629,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.06827011704444885,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 4.6158,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.08165498077869415,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 4.5782,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.06715558469295502,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 4.5527,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.08345365524291992,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 4.5388,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.11103973537683487,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 4.5174,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.08582151681184769,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 4.4972,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.08831771463155746,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 4.4765,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.08771367371082306,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 4.468,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.12151385098695755,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 4.4441,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.07800640165805817,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 4.4317,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.0812416672706604,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 4.4408,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.08056843280792236,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 4.4374,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.07432050257921219,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 4.4161,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.09560892730951309,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 4.3988,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.07960548996925354,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 4.4058,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.06768720597028732,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 4.3873,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.07363396137952805,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 4.3811,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.09380222111940384,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 4.3637,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.07489917427301407,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 4.3615,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.08118239790201187,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 4.3761,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.09006337821483612,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 4.3647,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.08469276875257492,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 4.3487,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.07916974276304245,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 4.3473,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.07793714106082916,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 4.3528,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.07320082932710648,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 4.3572,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.06270208954811096,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 4.3392,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.06513089686632156,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 4.3544,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.07348134368658066,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 4.3385,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.07335767894983292,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 4.3377,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.06236434727907181,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 4.3372,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.07267049700021744,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 4.3295,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.06093907356262207,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 4.3339,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.07383224368095398,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 4.3412,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.06044156476855278,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 4.3336,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.06623392552137375,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 4.3248,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.06142057850956917,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 4.3119,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.06617292016744614,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 4.3227,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.06499040126800537,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 4.3219,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.05966311693191528,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 4.3377,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.0569876991212368,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 4.3245,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.06260077655315399,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 4.3178,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.056433532387018204,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 4.3199,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.058227624744176865,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 4.3236,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 3.289541162754376e+17,
|
| 394 |
+
"train_loss": 4.664218570114276,
|
| 395 |
+
"train_runtime": 5076.8146,
|
| 396 |
+
"train_samples_per_second": 109.937,
|
| 397 |
+
"train_steps_per_second": 0.107
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 31896544,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 3.289541162754376e+17,
|
| 418 |
+
"train_batch_size": 32,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen3/eupe_convnext_b/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
continuous/qwen3/eupe_convnext_b/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|
continuous/qwen3/eupe_convnext_b/finetune/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaQwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 1024,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": 0.0001,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter,mm_language_model",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ConvNeXt-B/EUPE-ConvNeXt-B.pt",
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/eupe_convnext_b/finetune/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_sample": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
151645,
|
| 5 |
+
151643
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 151643,
|
| 8 |
+
"temperature": 0.6,
|
| 9 |
+
"top_k": 20,
|
| 10 |
+
"top_p": 0.95,
|
| 11 |
+
"transformers_version": "4.57.6"
|
| 12 |
+
}
|
continuous/qwen3/eupe_convnext_b/finetune/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/eupe_convnext_b/finetune/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:660dea7df0d0841ef7c31656bffffdd190470dd54669cf9fdf0cc423d70c355f
|
| 3 |
+
size 4263879864
|
continuous/qwen3/eupe_convnext_b/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/qwen3/eupe_convnext_b/finetune/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
continuous/qwen3/eupe_convnext_b/finetune/tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"clean_up_tokenization_spaces": false,
|
| 231 |
+
"eos_token": "<|im_end|>",
|
| 232 |
+
"errors": "replace",
|
| 233 |
+
"extra_special_tokens": {},
|
| 234 |
+
"model_max_length": 8192,
|
| 235 |
+
"pad_token": "<|endoftext|>",
|
| 236 |
+
"padding_side": "right",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
continuous/qwen3/eupe_convnext_b/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": 17.710569381713867,
|
| 15 |
+
"learning_rate": 1.153846153846154e-06,
|
| 16 |
+
"loss": 2.6165,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.007700452401578592,
|
| 21 |
+
"grad_norm": 7.226413726806641,
|
| 22 |
+
"learning_rate": 2.435897435897436e-06,
|
| 23 |
+
"loss": 2.0615,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01155067860236789,
|
| 28 |
+
"grad_norm": 4.506906509399414,
|
| 29 |
+
"learning_rate": 3.7179487179487184e-06,
|
| 30 |
+
"loss": 1.5354,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015400904803157185,
|
| 35 |
+
"grad_norm": 1.435056447982788,
|
| 36 |
+
"learning_rate": 5e-06,
|
| 37 |
+
"loss": 1.195,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.019251131003946483,
|
| 42 |
+
"grad_norm": 0.7689356207847595,
|
| 43 |
+
"learning_rate": 6.282051282051282e-06,
|
| 44 |
+
"loss": 1.0802,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02310135720473578,
|
| 49 |
+
"grad_norm": 0.4086686074733734,
|
| 50 |
+
"learning_rate": 7.564102564102564e-06,
|
| 51 |
+
"loss": 0.9917,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.026951583405525073,
|
| 56 |
+
"grad_norm": 0.3145039677619934,
|
| 57 |
+
"learning_rate": 8.846153846153847e-06,
|
| 58 |
+
"loss": 0.9413,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03080180960631437,
|
| 63 |
+
"grad_norm": 0.3004060685634613,
|
| 64 |
+
"learning_rate": 9.999996111488723e-06,
|
| 65 |
+
"loss": 0.9371,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.034652035807103666,
|
| 70 |
+
"grad_norm": 0.2833903729915619,
|
| 71 |
+
"learning_rate": 9.999529497453782e-06,
|
| 72 |
+
"loss": 0.9032,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.038502262007892966,
|
| 77 |
+
"grad_norm": 0.25233349204063416,
|
| 78 |
+
"learning_rate": 9.998285264324193e-06,
|
| 79 |
+
"loss": 0.8858,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04235248820868226,
|
| 84 |
+
"grad_norm": 0.2690836191177368,
|
| 85 |
+
"learning_rate": 9.996263605626052e-06,
|
| 86 |
+
"loss": 0.8661,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04620271440947156,
|
| 91 |
+
"grad_norm": 0.2487419843673706,
|
| 92 |
+
"learning_rate": 9.99346483580503e-06,
|
| 93 |
+
"loss": 0.8756,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05005294061026085,
|
| 98 |
+
"grad_norm": 0.2509314715862274,
|
| 99 |
+
"learning_rate": 9.989889390177467e-06,
|
| 100 |
+
"loss": 0.8724,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.053903166811050146,
|
| 105 |
+
"grad_norm": 0.28489118814468384,
|
| 106 |
+
"learning_rate": 9.985537824862647e-06,
|
| 107 |
+
"loss": 0.8662,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.057753393011839446,
|
| 112 |
+
"grad_norm": 0.29184389114379883,
|
| 113 |
+
"learning_rate": 9.980410816696317e-06,
|
| 114 |
+
"loss": 0.851,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06160361921262874,
|
| 119 |
+
"grad_norm": 0.25297704339027405,
|
| 120 |
+
"learning_rate": 9.97450916312541e-06,
|
| 121 |
+
"loss": 0.8554,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06545384541341803,
|
| 126 |
+
"grad_norm": 0.23005062341690063,
|
| 127 |
+
"learning_rate": 9.967833782084004e-06,
|
| 128 |
+
"loss": 0.8568,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.06930407161420733,
|
| 133 |
+
"grad_norm": 0.26323211193084717,
|
| 134 |
+
"learning_rate": 9.960385711850554e-06,
|
| 135 |
+
"loss": 0.8843,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07315429781499663,
|
| 140 |
+
"grad_norm": 0.25779634714126587,
|
| 141 |
+
"learning_rate": 9.952166110886394e-06,
|
| 142 |
+
"loss": 0.8731,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.07700452401578593,
|
| 147 |
+
"grad_norm": 0.27655893564224243,
|
| 148 |
+
"learning_rate": 9.943176257655567e-06,
|
| 149 |
+
"loss": 0.8633,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08085475021657522,
|
| 154 |
+
"grad_norm": 0.2567359209060669,
|
| 155 |
+
"learning_rate": 9.933417550425952e-06,
|
| 156 |
+
"loss": 0.8575,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08470497641736452,
|
| 161 |
+
"grad_norm": 0.25726014375686646,
|
| 162 |
+
"learning_rate": 9.9228915070518e-06,
|
| 163 |
+
"loss": 0.8453,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08855520261815382,
|
| 168 |
+
"grad_norm": 0.24708588421344757,
|
| 169 |
+
"learning_rate": 9.911599764737632e-06,
|
| 170 |
+
"loss": 0.8457,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.09240542881894312,
|
| 175 |
+
"grad_norm": 0.2525484263896942,
|
| 176 |
+
"learning_rate": 9.899544079783606e-06,
|
| 177 |
+
"loss": 0.841,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0962556550197324,
|
| 182 |
+
"grad_norm": 0.2965702414512634,
|
| 183 |
+
"learning_rate": 9.886726327312338e-06,
|
| 184 |
+
"loss": 0.8573,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1001058812205217,
|
| 189 |
+
"grad_norm": 0.24055618047714233,
|
| 190 |
+
"learning_rate": 9.87314850097724e-06,
|
| 191 |
+
"loss": 0.8543,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.103956107421311,
|
| 196 |
+
"grad_norm": 0.24427670240402222,
|
| 197 |
+
"learning_rate": 9.858812712652452e-06,
|
| 198 |
+
"loss": 0.842,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10780633362210029,
|
| 203 |
+
"grad_norm": 0.23616403341293335,
|
| 204 |
+
"learning_rate": 9.843721192104341e-06,
|
| 205 |
+
"loss": 0.8419,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11165655982288959,
|
| 210 |
+
"grad_norm": 0.24669721722602844,
|
| 211 |
+
"learning_rate": 9.827876286644698e-06,
|
| 212 |
+
"loss": 0.8594,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.11550678602367889,
|
| 217 |
+
"grad_norm": 0.22014838457107544,
|
| 218 |
+
"learning_rate": 9.811280460765643e-06,
|
| 219 |
+
"loss": 0.8193,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11935701222446819,
|
| 224 |
+
"grad_norm": 0.24254633486270905,
|
| 225 |
+
"learning_rate": 9.793936295756292e-06,
|
| 226 |
+
"loss": 0.8335,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.12320723842525748,
|
| 231 |
+
"grad_norm": 0.24373257160186768,
|
| 232 |
+
"learning_rate": 9.775846489301279e-06,
|
| 233 |
+
"loss": 0.8329,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.12705746462604678,
|
| 238 |
+
"grad_norm": 0.24751658737659454,
|
| 239 |
+
"learning_rate": 9.757013855061157e-06,
|
| 240 |
+
"loss": 0.8327,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13090769082683607,
|
| 245 |
+
"grad_norm": 0.24443672597408295,
|
| 246 |
+
"learning_rate": 9.737441322234758e-06,
|
| 247 |
+
"loss": 0.8447,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.13475791702762538,
|
| 252 |
+
"grad_norm": 0.2717336416244507,
|
| 253 |
+
"learning_rate": 9.717131935103605e-06,
|
| 254 |
+
"loss": 0.8433,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.13860814322841467,
|
| 259 |
+
"grad_norm": 0.236042782664299,
|
| 260 |
+
"learning_rate": 9.696088852558398e-06,
|
| 261 |
+
"loss": 0.8302,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.14245836942920398,
|
| 266 |
+
"grad_norm": 0.2516211271286011,
|
| 267 |
+
"learning_rate": 9.674315347607686e-06,
|
| 268 |
+
"loss": 0.8568,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.14630859562999327,
|
| 273 |
+
"grad_norm": 0.2312365472316742,
|
| 274 |
+
"learning_rate": 9.651814806868796e-06,
|
| 275 |
+
"loss": 0.8167,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15015882183078255,
|
| 280 |
+
"grad_norm": 0.24406522512435913,
|
| 281 |
+
"learning_rate": 9.62859073004107e-06,
|
| 282 |
+
"loss": 0.8254,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.15400904803157187,
|
| 287 |
+
"grad_norm": 0.23971618711948395,
|
| 288 |
+
"learning_rate": 9.604646729361547e-06,
|
| 289 |
+
"loss": 0.8278,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.15785927423236115,
|
| 294 |
+
"grad_norm": 0.2310664802789688,
|
| 295 |
+
"learning_rate": 9.579986529043099e-06,
|
| 296 |
+
"loss": 0.8162,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.16170950043315044,
|
| 301 |
+
"grad_norm": 0.2363699972629547,
|
| 302 |
+
"learning_rate": 9.554613964695189e-06,
|
| 303 |
+
"loss": 0.8327,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.16555972663393975,
|
| 308 |
+
"grad_norm": 0.24628883600234985,
|
| 309 |
+
"learning_rate": 9.528532982727281e-06,
|
| 310 |
+
"loss": 0.8218,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.16940995283472904,
|
| 315 |
+
"grad_norm": 0.24092219769954681,
|
| 316 |
+
"learning_rate": 9.501747639735018e-06,
|
| 317 |
+
"loss": 0.8197,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.17326017903551832,
|
| 322 |
+
"grad_norm": 0.25136205554008484,
|
| 323 |
+
"learning_rate": 9.474262101869272e-06,
|
| 324 |
+
"loss": 0.8095,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.17711040523630764,
|
| 329 |
+
"grad_norm": 0.2517487406730652,
|
| 330 |
+
"learning_rate": 9.446080644188141e-06,
|
| 331 |
+
"loss": 0.8148,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.18096063143709692,
|
| 336 |
+
"grad_norm": 0.24557827413082123,
|
| 337 |
+
"learning_rate": 9.417207649992017e-06,
|
| 338 |
+
"loss": 0.8237,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.18481085763788624,
|
| 343 |
+
"grad_norm": 0.24084416031837463,
|
| 344 |
+
"learning_rate": 9.387647610141816e-06,
|
| 345 |
+
"loss": 0.822,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18866108383867553,
|
| 350 |
+
"grad_norm": 0.2305324226617813,
|
| 351 |
+
"learning_rate": 9.35740512236047e-06,
|
| 352 |
+
"loss": 0.8083,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.1925113100394648,
|
| 357 |
+
"grad_norm": 0.22488975524902344,
|
| 358 |
+
"learning_rate": 9.326484890517809e-06,
|
| 359 |
+
"loss": 0.8131,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.19636153624025413,
|
| 364 |
+
"grad_norm": 0.2294781655073166,
|
| 365 |
+
"learning_rate": 9.294891723898926e-06,
|
| 366 |
+
"loss": 0.8093,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2002117624410434,
|
| 371 |
+
"grad_norm": 0.24125033617019653,
|
| 372 |
+
"learning_rate": 9.262630536456153e-06,
|
| 373 |
+
"loss": 0.8388,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2040619886418327,
|
| 378 |
+
"grad_norm": 0.24610374867916107,
|
| 379 |
+
"learning_rate": 9.229706346044749e-06,
|
| 380 |
+
"loss": 0.8104,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.207912214842622,
|
| 385 |
+
"grad_norm": 0.24861371517181396,
|
| 386 |
+
"learning_rate": 9.196124273642422e-06,
|
| 387 |
+
"loss": 0.8244,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2117624410434113,
|
| 392 |
+
"grad_norm": 0.227766752243042,
|
| 393 |
+
"learning_rate": 9.161889542552842e-06,
|
| 394 |
+
"loss": 0.8258,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.21561266724420058,
|
| 399 |
+
"grad_norm": 0.22121156752109528,
|
| 400 |
+
"learning_rate": 9.127007477593193e-06,
|
| 401 |
+
"loss": 0.8309,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2194628934449899,
|
| 406 |
+
"grad_norm": 0.26176249980926514,
|
| 407 |
+
"learning_rate": 9.091483504265974e-06,
|
| 408 |
+
"loss": 0.8393,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.22331311964577918,
|
| 413 |
+
"grad_norm": 0.25159117579460144,
|
| 414 |
+
"learning_rate": 9.055323147915116e-06,
|
| 415 |
+
"loss": 0.8194,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2271633458465685,
|
| 420 |
+
"grad_norm": 0.2524913251399994,
|
| 421 |
+
"learning_rate": 9.018532032866585e-06,
|
| 422 |
+
"loss": 0.8297,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.23101357204735778,
|
| 427 |
+
"grad_norm": 0.23200897872447968,
|
| 428 |
+
"learning_rate": 8.981115881553579e-06,
|
| 429 |
+
"loss": 0.821,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.23486379824814707,
|
| 434 |
+
"grad_norm": 0.254387766122818,
|
| 435 |
+
"learning_rate": 8.943080513626476e-06,
|
| 436 |
+
"loss": 0.8402,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.23871402444893638,
|
| 441 |
+
"grad_norm": 0.23568791151046753,
|
| 442 |
+
"learning_rate": 8.904431845047641e-06,
|
| 443 |
+
"loss": 0.8249,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.24256425064972567,
|
| 448 |
+
"grad_norm": 0.24557791650295258,
|
| 449 |
+
"learning_rate": 8.86517588717128e-06,
|
| 450 |
+
"loss": 0.8414,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.24641447685051496,
|
| 455 |
+
"grad_norm": 0.2321411818265915,
|
| 456 |
+
"learning_rate": 8.82531874580844e-06,
|
| 457 |
+
"loss": 0.8297,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.25026470305130427,
|
| 462 |
+
"grad_norm": 0.21997655928134918,
|
| 463 |
+
"learning_rate": 8.784866620277314e-06,
|
| 464 |
+
"loss": 0.7874,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.25411492925209356,
|
| 469 |
+
"grad_norm": 0.25585925579071045,
|
| 470 |
+
"learning_rate": 8.743825802439018e-06,
|
| 471 |
+
"loss": 0.8247,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.25796515545288284,
|
| 476 |
+
"grad_norm": 0.2669120728969574,
|
| 477 |
+
"learning_rate": 8.702202675718956e-06,
|
| 478 |
+
"loss": 0.8296,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.26181538165367213,
|
| 483 |
+
"grad_norm": 0.23826593160629272,
|
| 484 |
+
"learning_rate": 8.660003714113959e-06,
|
| 485 |
+
"loss": 0.7999,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.26566560785446147,
|
| 490 |
+
"grad_norm": 0.23396848142147064,
|
| 491 |
+
"learning_rate": 8.617235481185323e-06,
|
| 492 |
+
"loss": 0.8249,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.26951583405525076,
|
| 497 |
+
"grad_norm": 0.23888027667999268,
|
| 498 |
+
"learning_rate": 8.573904629037931e-06,
|
| 499 |
+
"loss": 0.8292,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.27336606025604004,
|
| 504 |
+
"grad_norm": 0.23711048066616058,
|
| 505 |
+
"learning_rate": 8.530017897285585e-06,
|
| 506 |
+
"loss": 0.8096,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.27721628645682933,
|
| 511 |
+
"grad_norm": 0.22162885963916779,
|
| 512 |
+
"learning_rate": 8.485582112002746e-06,
|
| 513 |
+
"loss": 0.8134,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2810665126576186,
|
| 518 |
+
"grad_norm": 0.23747703433036804,
|
| 519 |
+
"learning_rate": 8.440604184662811e-06,
|
| 520 |
+
"loss": 0.8195,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.28491673885840796,
|
| 525 |
+
"grad_norm": 0.23633407056331635,
|
| 526 |
+
"learning_rate": 8.395091111063118e-06,
|
| 527 |
+
"loss": 0.8081,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.28876696505919724,
|
| 532 |
+
"grad_norm": 0.21849820017814636,
|
| 533 |
+
"learning_rate": 8.349049970236822e-06,
|
| 534 |
+
"loss": 0.8072,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.29261719125998653,
|
| 539 |
+
"grad_norm": 0.23549100756645203,
|
| 540 |
+
"learning_rate": 8.302487923351842e-06,
|
| 541 |
+
"loss": 0.7998,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2964674174607758,
|
| 546 |
+
"grad_norm": 0.2311391532421112,
|
| 547 |
+
"learning_rate": 8.255412212597021e-06,
|
| 548 |
+
"loss": 0.8176,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3003176436615651,
|
| 553 |
+
"grad_norm": 0.2290457785129547,
|
| 554 |
+
"learning_rate": 8.207830160055688e-06,
|
| 555 |
+
"loss": 0.8142,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3041678698623544,
|
| 560 |
+
"grad_norm": 0.24196819961071014,
|
| 561 |
+
"learning_rate": 8.159749166566786e-06,
|
| 562 |
+
"loss": 0.799,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.30801809606314373,
|
| 567 |
+
"grad_norm": 0.22549355030059814,
|
| 568 |
+
"learning_rate": 8.11117671057377e-06,
|
| 569 |
+
"loss": 0.8267,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.311868322263933,
|
| 574 |
+
"grad_norm": 0.2442585676908493,
|
| 575 |
+
"learning_rate": 8.062120346961404e-06,
|
| 576 |
+
"loss": 0.8185,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3157185484647223,
|
| 581 |
+
"grad_norm": 0.2380821406841278,
|
| 582 |
+
"learning_rate": 8.012587705880705e-06,
|
| 583 |
+
"loss": 0.8114,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3195687746655116,
|
| 588 |
+
"grad_norm": 0.2348862886428833,
|
| 589 |
+
"learning_rate": 7.962586491562146e-06,
|
| 590 |
+
"loss": 0.7979,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3234190008663009,
|
| 595 |
+
"grad_norm": 0.2472226619720459,
|
| 596 |
+
"learning_rate": 7.912124481117357e-06,
|
| 597 |
+
"loss": 0.8041,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3272692270670902,
|
| 602 |
+
"grad_norm": 0.23739391565322876,
|
| 603 |
+
"learning_rate": 7.861209523329494e-06,
|
| 604 |
+
"loss": 0.814,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3311194532678795,
|
| 609 |
+
"grad_norm": 0.2286122590303421,
|
| 610 |
+
"learning_rate": 7.809849537432432e-06,
|
| 611 |
+
"loss": 0.7981,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3349696794686688,
|
| 616 |
+
"grad_norm": 0.23188607394695282,
|
| 617 |
+
"learning_rate": 7.758052511879035e-06,
|
| 618 |
+
"loss": 0.8082,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3388199056694581,
|
| 623 |
+
"grad_norm": 0.21374401450157166,
|
| 624 |
+
"learning_rate": 7.705826503098642e-06,
|
| 625 |
+
"loss": 0.8028,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.34267013187024736,
|
| 630 |
+
"grad_norm": 0.22572943568229675,
|
| 631 |
+
"learning_rate": 7.653179634243979e-06,
|
| 632 |
+
"loss": 0.8032,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.34652035807103665,
|
| 637 |
+
"grad_norm": 0.2234373241662979,
|
| 638 |
+
"learning_rate": 7.600120093927696e-06,
|
| 639 |
+
"loss": 0.809,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.350370584271826,
|
| 644 |
+
"grad_norm": 0.24654211103916168,
|
| 645 |
+
"learning_rate": 7.546656134948729e-06,
|
| 646 |
+
"loss": 0.7992,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3542208104726153,
|
| 651 |
+
"grad_norm": 0.23832322657108307,
|
| 652 |
+
"learning_rate": 7.492796073008661e-06,
|
| 653 |
+
"loss": 0.7969,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.35807103667340456,
|
| 658 |
+
"grad_norm": 0.25864624977111816,
|
| 659 |
+
"learning_rate": 7.438548285418326e-06,
|
| 660 |
+
"loss": 0.7826,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.36192126287419385,
|
| 665 |
+
"grad_norm": 0.2499522715806961,
|
| 666 |
+
"learning_rate": 7.3839212097948e-06,
|
| 667 |
+
"loss": 0.8169,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.36577148907498314,
|
| 672 |
+
"grad_norm": 0.23514677584171295,
|
| 673 |
+
"learning_rate": 7.328923342749039e-06,
|
| 674 |
+
"loss": 0.8317,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3696217152757725,
|
| 679 |
+
"grad_norm": 0.25035005807876587,
|
| 680 |
+
"learning_rate": 7.273563238564319e-06,
|
| 681 |
+
"loss": 0.8156,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.37347194147656176,
|
| 686 |
+
"grad_norm": 0.23709125816822052,
|
| 687 |
+
"learning_rate": 7.217849507865724e-06,
|
| 688 |
+
"loss": 0.7973,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.37732216767735105,
|
| 693 |
+
"grad_norm": 0.22676785290241241,
|
| 694 |
+
"learning_rate": 7.161790816280858e-06,
|
| 695 |
+
"loss": 0.8016,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.38117239387814034,
|
| 700 |
+
"grad_norm": 0.23334160447120667,
|
| 701 |
+
"learning_rate": 7.1053958830920125e-06,
|
| 702 |
+
"loss": 0.7895,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3850226200789296,
|
| 707 |
+
"grad_norm": 0.23835702240467072,
|
| 708 |
+
"learning_rate": 7.0486734798799715e-06,
|
| 709 |
+
"loss": 0.8041,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3888728462797189,
|
| 714 |
+
"grad_norm": 0.2409580498933792,
|
| 715 |
+
"learning_rate": 6.991632429159708e-06,
|
| 716 |
+
"loss": 0.8278,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.39272307248050825,
|
| 721 |
+
"grad_norm": 0.24837662279605865,
|
| 722 |
+
"learning_rate": 6.934281603008132e-06,
|
| 723 |
+
"loss": 0.8196,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.39657329868129754,
|
| 728 |
+
"grad_norm": 0.23517023026943207,
|
| 729 |
+
"learning_rate": 6.876629921684143e-06,
|
| 730 |
+
"loss": 0.8164,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4004235248820868,
|
| 735 |
+
"grad_norm": 0.241703063249588,
|
| 736 |
+
"learning_rate": 6.818686352241199e-06,
|
| 737 |
+
"loss": 0.8073,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4042737510828761,
|
| 742 |
+
"grad_norm": 0.22045741975307465,
|
| 743 |
+
"learning_rate": 6.760459907132573e-06,
|
| 744 |
+
"loss": 0.8033,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.4081239772836654,
|
| 749 |
+
"grad_norm": 0.22270038723945618,
|
| 750 |
+
"learning_rate": 6.701959642809589e-06,
|
| 751 |
+
"loss": 0.8105,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.41197420348445474,
|
| 756 |
+
"grad_norm": 0.23214313387870789,
|
| 757 |
+
"learning_rate": 6.643194658312987e-06,
|
| 758 |
+
"loss": 0.8024,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.415824429685244,
|
| 763 |
+
"grad_norm": 0.23270408809185028,
|
| 764 |
+
"learning_rate": 6.584174093857676e-06,
|
| 765 |
+
"loss": 0.8261,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4196746558860333,
|
| 770 |
+
"grad_norm": 0.25575587153434753,
|
| 771 |
+
"learning_rate": 6.5249071294110735e-06,
|
| 772 |
+
"loss": 0.8058,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.4235248820868226,
|
| 777 |
+
"grad_norm": 0.25121214985847473,
|
| 778 |
+
"learning_rate": 6.465402983265274e-06,
|
| 779 |
+
"loss": 0.8206,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4273751082876119,
|
| 784 |
+
"grad_norm": 0.2412593960762024,
|
| 785 |
+
"learning_rate": 6.405670910603243e-06,
|
| 786 |
+
"loss": 0.8162,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.43122533448840117,
|
| 791 |
+
"grad_norm": 0.25452354550361633,
|
| 792 |
+
"learning_rate": 6.345720202059291e-06,
|
| 793 |
+
"loss": 0.8122,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4350755606891905,
|
| 798 |
+
"grad_norm": 0.23888421058654785,
|
| 799 |
+
"learning_rate": 6.285560182274008e-06,
|
| 800 |
+
"loss": 0.791,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4389257868899798,
|
| 805 |
+
"grad_norm": 0.2358405441045761,
|
| 806 |
+
"learning_rate": 6.225200208443936e-06,
|
| 807 |
+
"loss": 0.8137,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.4427760130907691,
|
| 812 |
+
"grad_norm": 0.2180458903312683,
|
| 813 |
+
"learning_rate": 6.16464966886616e-06,
|
| 814 |
+
"loss": 0.8038,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.44662623929155837,
|
| 819 |
+
"grad_norm": 0.26270103454589844,
|
| 820 |
+
"learning_rate": 6.103917981478065e-06,
|
| 821 |
+
"loss": 0.8262,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.45047646549234766,
|
| 826 |
+
"grad_norm": 0.22341342270374298,
|
| 827 |
+
"learning_rate": 6.043014592392489e-06,
|
| 828 |
+
"loss": 0.8375,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.454326691693137,
|
| 833 |
+
"grad_norm": 0.22343017160892487,
|
| 834 |
+
"learning_rate": 5.981948974428485e-06,
|
| 835 |
+
"loss": 0.8127,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.4581769178939263,
|
| 840 |
+
"grad_norm": 0.24608784914016724,
|
| 841 |
+
"learning_rate": 5.920730625637934e-06,
|
| 842 |
+
"loss": 0.7927,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.46202714409471557,
|
| 847 |
+
"grad_norm": 0.24347972869873047,
|
| 848 |
+
"learning_rate": 5.8593690678282355e-06,
|
| 849 |
+
"loss": 0.8015,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.46587737029550486,
|
| 854 |
+
"grad_norm": 0.24312902987003326,
|
| 855 |
+
"learning_rate": 5.797873845081301e-06,
|
| 856 |
+
"loss": 0.8062,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.46972759649629414,
|
| 861 |
+
"grad_norm": 0.23362381756305695,
|
| 862 |
+
"learning_rate": 5.736254522269075e-06,
|
| 863 |
+
"loss": 0.8229,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.47357782269708343,
|
| 868 |
+
"grad_norm": 0.22823509573936462,
|
| 869 |
+
"learning_rate": 5.6745206835658386e-06,
|
| 870 |
+
"loss": 0.824,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.47742804889787277,
|
| 875 |
+
"grad_norm": 0.23175545036792755,
|
| 876 |
+
"learning_rate": 5.612681930957495e-06,
|
| 877 |
+
"loss": 0.7947,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.48127827509866206,
|
| 882 |
+
"grad_norm": 0.2197483330965042,
|
| 883 |
+
"learning_rate": 5.550747882748095e-06,
|
| 884 |
+
"loss": 0.7945,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.48512850129945134,
|
| 889 |
+
"grad_norm": 0.2461317926645279,
|
| 890 |
+
"learning_rate": 5.4887281720638195e-06,
|
| 891 |
+
"loss": 0.7988,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.48897872750024063,
|
| 896 |
+
"grad_norm": 0.2709513008594513,
|
| 897 |
+
"learning_rate": 5.42663244535466e-06,
|
| 898 |
+
"loss": 0.7878,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4928289537010299,
|
| 903 |
+
"grad_norm": 0.25671449303627014,
|
| 904 |
+
"learning_rate": 5.3644703608940205e-06,
|
| 905 |
+
"loss": 0.7936,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.49667917990181926,
|
| 910 |
+
"grad_norm": 0.23091107606887817,
|
| 911 |
+
"learning_rate": 5.302251587276491e-06,
|
| 912 |
+
"loss": 0.805,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5005294061026085,
|
| 917 |
+
"grad_norm": 0.22907423973083496,
|
| 918 |
+
"learning_rate": 5.2399858019140005e-06,
|
| 919 |
+
"loss": 0.8047,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5043796323033978,
|
| 924 |
+
"grad_norm": 0.22674375772476196,
|
| 925 |
+
"learning_rate": 5.177682689530615e-06,
|
| 926 |
+
"loss": 0.8194,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5082298585041871,
|
| 931 |
+
"grad_norm": 0.23817230761051178,
|
| 932 |
+
"learning_rate": 5.11535194065619e-06,
|
| 933 |
+
"loss": 0.7868,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5120800847049765,
|
| 938 |
+
"grad_norm": 0.2549503743648529,
|
| 939 |
+
"learning_rate": 5.053003250119113e-06,
|
| 940 |
+
"loss": 0.7991,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5159303109057657,
|
| 945 |
+
"grad_norm": 0.23804745078086853,
|
| 946 |
+
"learning_rate": 4.9906463155383925e-06,
|
| 947 |
+
"loss": 0.8048,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.519780537106555,
|
| 952 |
+
"grad_norm": 0.2319086790084839,
|
| 953 |
+
"learning_rate": 4.9282908358153046e-06,
|
| 954 |
+
"loss": 0.7956,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5236307633073443,
|
| 959 |
+
"grad_norm": 0.25518104434013367,
|
| 960 |
+
"learning_rate": 4.8659465096248345e-06,
|
| 961 |
+
"loss": 0.8149,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5274809895081336,
|
| 966 |
+
"grad_norm": 0.2620072662830353,
|
| 967 |
+
"learning_rate": 4.803623033907169e-06,
|
| 968 |
+
"loss": 0.8147,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5313312157089229,
|
| 973 |
+
"grad_norm": 0.2370920330286026,
|
| 974 |
+
"learning_rate": 4.741330102359441e-06,
|
| 975 |
+
"loss": 0.819,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5351814419097122,
|
| 980 |
+
"grad_norm": 0.24922727048397064,
|
| 981 |
+
"learning_rate": 4.679077403927996e-06,
|
| 982 |
+
"loss": 0.7993,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5390316681105015,
|
| 987 |
+
"grad_norm": 0.22953003644943237,
|
| 988 |
+
"learning_rate": 4.616874621301373e-06,
|
| 989 |
+
"loss": 0.7987,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5428818943112907,
|
| 994 |
+
"grad_norm": 0.30042147636413574,
|
| 995 |
+
"learning_rate": 4.554731429404293e-06,
|
| 996 |
+
"loss": 0.8134,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5467321205120801,
|
| 1001 |
+
"grad_norm": 0.2541881501674652,
|
| 1002 |
+
"learning_rate": 4.492657493892822e-06,
|
| 1003 |
+
"loss": 0.7995,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5505823467128694,
|
| 1008 |
+
"grad_norm": 0.26170021295547485,
|
| 1009 |
+
"learning_rate": 4.430662469650999e-06,
|
| 1010 |
+
"loss": 0.8138,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5544325729136587,
|
| 1015 |
+
"grad_norm": 0.25961658358573914,
|
| 1016 |
+
"learning_rate": 4.368755999289118e-06,
|
| 1017 |
+
"loss": 0.8158,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.558282799114448,
|
| 1022 |
+
"grad_norm": 0.26379916071891785,
|
| 1023 |
+
"learning_rate": 4.306947711643949e-06,
|
| 1024 |
+
"loss": 0.8022,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5621330253152372,
|
| 1029 |
+
"grad_norm": 0.24228999018669128,
|
| 1030 |
+
"learning_rate": 4.245247220281068e-06,
|
| 1031 |
+
"loss": 0.8041,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5659832515160266,
|
| 1036 |
+
"grad_norm": 0.23579835891723633,
|
| 1037 |
+
"learning_rate": 4.183664121999587e-06,
|
| 1038 |
+
"loss": 0.8024,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.5698334777168159,
|
| 1043 |
+
"grad_norm": 0.2506634294986725,
|
| 1044 |
+
"learning_rate": 4.122207995339474e-06,
|
| 1045 |
+
"loss": 0.7974,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.5736837039176051,
|
| 1050 |
+
"grad_norm": 0.22962439060211182,
|
| 1051 |
+
"learning_rate": 4.060888399091734e-06,
|
| 1052 |
+
"loss": 0.804,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.5775339301183945,
|
| 1057 |
+
"grad_norm": 0.22841216623783112,
|
| 1058 |
+
"learning_rate": 3.999714870811639e-06,
|
| 1059 |
+
"loss": 0.811,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5813841563191837,
|
| 1064 |
+
"grad_norm": 0.2340562343597412,
|
| 1065 |
+
"learning_rate": 3.938696925335275e-06,
|
| 1066 |
+
"loss": 0.8051,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.5852343825199731,
|
| 1071 |
+
"grad_norm": 0.23778726160526276,
|
| 1072 |
+
"learning_rate": 3.87784405329962e-06,
|
| 1073 |
+
"loss": 0.8122,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.5890846087207623,
|
| 1078 |
+
"grad_norm": 0.23762060701847076,
|
| 1079 |
+
"learning_rate": 3.817165719666382e-06,
|
| 1080 |
+
"loss": 0.8162,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.5929348349215516,
|
| 1085 |
+
"grad_norm": 0.22762183845043182,
|
| 1086 |
+
"learning_rate": 3.756671362249841e-06,
|
| 1087 |
+
"loss": 0.8179,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.596785061122341,
|
| 1092 |
+
"grad_norm": 0.23682498931884766,
|
| 1093 |
+
"learning_rate": 3.6963703902488907e-06,
|
| 1094 |
+
"loss": 0.8055,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6006352873231302,
|
| 1099 |
+
"grad_norm": 0.22838838398456573,
|
| 1100 |
+
"learning_rate": 3.6362721827835655e-06,
|
| 1101 |
+
"loss": 0.8244,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6044855135239195,
|
| 1106 |
+
"grad_norm": 0.24817195534706116,
|
| 1107 |
+
"learning_rate": 3.576386087436212e-06,
|
| 1108 |
+
"loss": 0.8173,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6083357397247088,
|
| 1113 |
+
"grad_norm": 0.24841882288455963,
|
| 1114 |
+
"learning_rate": 3.5167214187975913e-06,
|
| 1115 |
+
"loss": 0.8075,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6121859659254981,
|
| 1120 |
+
"grad_norm": 0.20251992344856262,
|
| 1121 |
+
"learning_rate": 3.457287457018086e-06,
|
| 1122 |
+
"loss": 0.8095,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6160361921262875,
|
| 1127 |
+
"grad_norm": 0.23062966763973236,
|
| 1128 |
+
"learning_rate": 3.3980934463643003e-06,
|
| 1129 |
+
"loss": 0.7964,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6198864183270767,
|
| 1134 |
+
"grad_norm": 0.22252106666564941,
|
| 1135 |
+
"learning_rate": 3.3391485937812053e-06,
|
| 1136 |
+
"loss": 0.7795,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.623736644527866,
|
| 1141 |
+
"grad_norm": 0.23149777948856354,
|
| 1142 |
+
"learning_rate": 3.280462067460115e-06,
|
| 1143 |
+
"loss": 0.7788,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6275868707286553,
|
| 1148 |
+
"grad_norm": 0.2329091578722,
|
| 1149 |
+
"learning_rate": 3.222042995412669e-06,
|
| 1150 |
+
"loss": 0.8094,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6314370969294446,
|
| 1155 |
+
"grad_norm": 0.21773508191108704,
|
| 1156 |
+
"learning_rate": 3.163900464051089e-06,
|
| 1157 |
+
"loss": 0.799,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.635287323130234,
|
| 1162 |
+
"grad_norm": 0.24301952123641968,
|
| 1163 |
+
"learning_rate": 3.1060435167748826e-06,
|
| 1164 |
+
"loss": 0.8077,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6391375493310232,
|
| 1169 |
+
"grad_norm": 0.23073157668113708,
|
| 1170 |
+
"learning_rate": 3.048481152564253e-06,
|
| 1171 |
+
"loss": 0.7886,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6429877755318125,
|
| 1176 |
+
"grad_norm": 0.2415754497051239,
|
| 1177 |
+
"learning_rate": 2.9912223245804017e-06,
|
| 1178 |
+
"loss": 0.8096,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6468380017326018,
|
| 1183 |
+
"grad_norm": 0.24210433661937714,
|
| 1184 |
+
"learning_rate": 2.934275938772981e-06,
|
| 1185 |
+
"loss": 0.8082,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6506882279333911,
|
| 1190 |
+
"grad_norm": 0.25920742750167847,
|
| 1191 |
+
"learning_rate": 2.8776508524948676e-06,
|
| 1192 |
+
"loss": 0.7974,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6545384541341804,
|
| 1197 |
+
"grad_norm": 0.2135160267353058,
|
| 1198 |
+
"learning_rate": 2.821355873124503e-06,
|
| 1199 |
+
"loss": 0.796,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6583886803349697,
|
| 1204 |
+
"grad_norm": 0.24727436900138855,
|
| 1205 |
+
"learning_rate": 2.7653997566960144e-06,
|
| 1206 |
+
"loss": 0.8097,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.662238906535759,
|
| 1211 |
+
"grad_norm": 0.2517285943031311,
|
| 1212 |
+
"learning_rate": 2.709791206537309e-06,
|
| 1213 |
+
"loss": 0.7997,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6660891327365482,
|
| 1218 |
+
"grad_norm": 0.24009212851524353,
|
| 1219 |
+
"learning_rate": 2.6545388719163747e-06,
|
| 1220 |
+
"loss": 0.8076,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.6699393589373376,
|
| 1225 |
+
"grad_norm": 0.23167411983013153,
|
| 1226 |
+
"learning_rate": 2.599651346695979e-06,
|
| 1227 |
+
"loss": 0.8042,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.6737895851381268,
|
| 1232 |
+
"grad_norm": 0.23864354193210602,
|
| 1233 |
+
"learning_rate": 2.545137167997007e-06,
|
| 1234 |
+
"loss": 0.8,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.6776398113389162,
|
| 1239 |
+
"grad_norm": 0.22101223468780518,
|
| 1240 |
+
"learning_rate": 2.4910048148705972e-06,
|
| 1241 |
+
"loss": 0.7889,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.6814900375397055,
|
| 1246 |
+
"grad_norm": 0.2461836189031601,
|
| 1247 |
+
"learning_rate": 2.4372627069793327e-06,
|
| 1248 |
+
"loss": 0.8313,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.6853402637404947,
|
| 1253 |
+
"grad_norm": 0.23404143750667572,
|
| 1254 |
+
"learning_rate": 2.3839192032876486e-06,
|
| 1255 |
+
"loss": 0.8181,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.6891904899412841,
|
| 1260 |
+
"grad_norm": 0.2291279137134552,
|
| 1261 |
+
"learning_rate": 2.3309826007617047e-06,
|
| 1262 |
+
"loss": 0.8286,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6930407161420733,
|
| 1267 |
+
"grad_norm": 0.2325659990310669,
|
| 1268 |
+
"learning_rate": 2.278461133078881e-06,
|
| 1269 |
+
"loss": 0.792,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6968909423428626,
|
| 1274 |
+
"grad_norm": 0.23446398973464966,
|
| 1275 |
+
"learning_rate": 2.2263629693471258e-06,
|
| 1276 |
+
"loss": 0.7829,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.700741168543652,
|
| 1281 |
+
"grad_norm": 0.23067206144332886,
|
| 1282 |
+
"learning_rate": 2.1746962128343414e-06,
|
| 1283 |
+
"loss": 0.8088,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7045913947444412,
|
| 1288 |
+
"grad_norm": 0.2363605946302414,
|
| 1289 |
+
"learning_rate": 2.1234688997080183e-06,
|
| 1290 |
+
"loss": 0.7962,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7084416209452306,
|
| 1295 |
+
"grad_norm": 0.24311144649982452,
|
| 1296 |
+
"learning_rate": 2.0726889977853053e-06,
|
| 1297 |
+
"loss": 0.8005,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7122918471460198,
|
| 1302 |
+
"grad_norm": 0.25065121054649353,
|
| 1303 |
+
"learning_rate": 2.022364405293703e-06,
|
| 1304 |
+
"loss": 0.8135,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7161420733468091,
|
| 1309 |
+
"grad_norm": 0.2269001305103302,
|
| 1310 |
+
"learning_rate": 1.9725029496425814e-06,
|
| 1311 |
+
"loss": 0.8051,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7199922995475985,
|
| 1316 |
+
"grad_norm": 0.24161072075366974,
|
| 1317 |
+
"learning_rate": 1.923112386205727e-06,
|
| 1318 |
+
"loss": 0.8037,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7238425257483877,
|
| 1323 |
+
"grad_norm": 0.2168298065662384,
|
| 1324 |
+
"learning_rate": 1.874200397115079e-06,
|
| 1325 |
+
"loss": 0.8092,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.727692751949177,
|
| 1330 |
+
"grad_norm": 0.24287842214107513,
|
| 1331 |
+
"learning_rate": 1.825774590065857e-06,
|
| 1332 |
+
"loss": 0.7944,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7315429781499663,
|
| 1337 |
+
"grad_norm": 0.2435370683670044,
|
| 1338 |
+
"learning_rate": 1.7778424971332813e-06,
|
| 1339 |
+
"loss": 0.8006,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7353932043507556,
|
| 1344 |
+
"grad_norm": 0.22916975617408752,
|
| 1345 |
+
"learning_rate": 1.7304115736010407e-06,
|
| 1346 |
+
"loss": 0.8004,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.739243430551545,
|
| 1351 |
+
"grad_norm": 0.2319374978542328,
|
| 1352 |
+
"learning_rate": 1.6834891968017108e-06,
|
| 1353 |
+
"loss": 0.7917,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7430936567523342,
|
| 1358 |
+
"grad_norm": 0.2427685558795929,
|
| 1359 |
+
"learning_rate": 1.6370826649692884e-06,
|
| 1360 |
+
"loss": 0.8138,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7469438829531235,
|
| 1365 |
+
"grad_norm": 0.23775331676006317,
|
| 1366 |
+
"learning_rate": 1.5911991961040463e-06,
|
| 1367 |
+
"loss": 0.8196,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7507941091539128,
|
| 1372 |
+
"grad_norm": 0.2375352680683136,
|
| 1373 |
+
"learning_rate": 1.545845926849851e-06,
|
| 1374 |
+
"loss": 0.8054,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7546443353547021,
|
| 1379 |
+
"grad_norm": 0.25188958644866943,
|
| 1380 |
+
"learning_rate": 1.5010299113841397e-06,
|
| 1381 |
+
"loss": 0.8237,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.7584945615554913,
|
| 1386 |
+
"grad_norm": 0.24514393508434296,
|
| 1387 |
+
"learning_rate": 1.4567581203207238e-06,
|
| 1388 |
+
"loss": 0.8115,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.7623447877562807,
|
| 1393 |
+
"grad_norm": 0.24682281911373138,
|
| 1394 |
+
"learning_rate": 1.413037439625593e-06,
|
| 1395 |
+
"loss": 0.8117,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.76619501395707,
|
| 1400 |
+
"grad_norm": 0.2480047643184662,
|
| 1401 |
+
"learning_rate": 1.3698746695458847e-06,
|
| 1402 |
+
"loss": 0.8046,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.7700452401578592,
|
| 1407 |
+
"grad_norm": 0.22645168006420135,
|
| 1408 |
+
"learning_rate": 1.3272765235521745e-06,
|
| 1409 |
+
"loss": 0.7961,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.7738954663586486,
|
| 1414 |
+
"grad_norm": 0.2226949781179428,
|
| 1415 |
+
"learning_rate": 1.2852496272942877e-06,
|
| 1416 |
+
"loss": 0.8033,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7777456925594378,
|
| 1421 |
+
"grad_norm": 0.22286657989025116,
|
| 1422 |
+
"learning_rate": 1.2438005175707396e-06,
|
| 1423 |
+
"loss": 0.792,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7815959187602272,
|
| 1428 |
+
"grad_norm": 0.22976166009902954,
|
| 1429 |
+
"learning_rate": 1.2029356413120297e-06,
|
| 1430 |
+
"loss": 0.779,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7854461449610165,
|
| 1435 |
+
"grad_norm": 0.22884294390678406,
|
| 1436 |
+
"learning_rate": 1.1626613545778786e-06,
|
| 1437 |
+
"loss": 0.7958,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.7892963711618057,
|
| 1442 |
+
"grad_norm": 0.21980871260166168,
|
| 1443 |
+
"learning_rate": 1.1229839215686294e-06,
|
| 1444 |
+
"loss": 0.7992,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.7931465973625951,
|
| 1449 |
+
"grad_norm": 0.2149050533771515,
|
| 1450 |
+
"learning_rate": 1.0839095136509188e-06,
|
| 1451 |
+
"loss": 0.7867,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.7969968235633843,
|
| 1456 |
+
"grad_norm": 0.2579258382320404,
|
| 1457 |
+
"learning_rate": 1.045444208397791e-06,
|
| 1458 |
+
"loss": 0.8086,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8008470497641736,
|
| 1463 |
+
"grad_norm": 0.23153464496135712,
|
| 1464 |
+
"learning_rate": 1.007593988643402e-06,
|
| 1465 |
+
"loss": 0.7865,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.804697275964963,
|
| 1470 |
+
"grad_norm": 0.22876876592636108,
|
| 1471 |
+
"learning_rate": 9.703647415524613e-07,
|
| 1472 |
+
"loss": 0.8091,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8085475021657522,
|
| 1477 |
+
"grad_norm": 0.23696456849575043,
|
| 1478 |
+
"learning_rate": 9.337622577045485e-07,
|
| 1479 |
+
"loss": 0.7981,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8123977283665416,
|
| 1484 |
+
"grad_norm": 0.22904109954833984,
|
| 1485 |
+
"learning_rate": 8.977922301934588e-07,
|
| 1486 |
+
"loss": 0.8055,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8162479545673308,
|
| 1491 |
+
"grad_norm": 0.23177917301654816,
|
| 1492 |
+
"learning_rate": 8.624602537417009e-07,
|
| 1493 |
+
"loss": 0.8087,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8200981807681201,
|
| 1498 |
+
"grad_norm": 0.24076442420482635,
|
| 1499 |
+
"learning_rate": 8.277718238303123e-07,
|
| 1500 |
+
"loss": 0.7973,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8239484069689095,
|
| 1505 |
+
"grad_norm": 0.23266002535820007,
|
| 1506 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1507 |
+
"loss": 0.7998,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8277986331696987,
|
| 1512 |
+
"grad_norm": 0.24205608665943146,
|
| 1513 |
+
"learning_rate": 7.603470842324223e-07,
|
| 1514 |
+
"loss": 0.793,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.831648859370488,
|
| 1519 |
+
"grad_norm": 0.22166137397289276,
|
| 1520 |
+
"learning_rate": 7.276212616857575e-07,
|
| 1521 |
+
"loss": 0.8089,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8354990855712773,
|
| 1526 |
+
"grad_norm": 0.22360549867153168,
|
| 1527 |
+
"learning_rate": 6.955599583279837e-07,
|
| 1528 |
+
"loss": 0.8108,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8393493117720666,
|
| 1533 |
+
"grad_norm": 0.2330475002527237,
|
| 1534 |
+
"learning_rate": 6.641681609246981e-07,
|
| 1535 |
+
"loss": 0.7999,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8431995379728559,
|
| 1540 |
+
"grad_norm": 0.22636352479457855,
|
| 1541 |
+
"learning_rate": 6.334507521075711e-07,
|
| 1542 |
+
"loss": 0.8097,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8470497641736452,
|
| 1547 |
+
"grad_norm": 0.2536192238330841,
|
| 1548 |
+
"learning_rate": 6.034125096149179e-07,
|
| 1549 |
+
"loss": 0.8009,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8508999903744345,
|
| 1554 |
+
"grad_norm": 0.24580535292625427,
|
| 1555 |
+
"learning_rate": 5.740581055485722e-07,
|
| 1556 |
+
"loss": 0.7745,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.8547502165752238,
|
| 1561 |
+
"grad_norm": 0.2160973995923996,
|
| 1562 |
+
"learning_rate": 5.453921056471961e-07,
|
| 1563 |
+
"loss": 0.8087,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.8586004427760131,
|
| 1568 |
+
"grad_norm": 0.24418912827968597,
|
| 1569 |
+
"learning_rate": 5.174189685761288e-07,
|
| 1570 |
+
"loss": 0.8006,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8624506689768023,
|
| 1575 |
+
"grad_norm": 0.2283034324645996,
|
| 1576 |
+
"learning_rate": 4.901430452338979e-07,
|
| 1577 |
+
"loss": 0.812,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8663008951775917,
|
| 1582 |
+
"grad_norm": 0.23039308190345764,
|
| 1583 |
+
"learning_rate": 4.6356857807548286e-07,
|
| 1584 |
+
"loss": 0.797,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.870151121378381,
|
| 1589 |
+
"grad_norm": 0.233271986246109,
|
| 1590 |
+
"learning_rate": 4.376997004524541e-07,
|
| 1591 |
+
"loss": 0.7864,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.8740013475791703,
|
| 1596 |
+
"grad_norm": 0.23586930334568024,
|
| 1597 |
+
"learning_rate": 4.12540435970073e-07,
|
| 1598 |
+
"loss": 0.7985,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.8778515737799596,
|
| 1603 |
+
"grad_norm": 0.22578252851963043,
|
| 1604 |
+
"learning_rate": 3.880946978614703e-07,
|
| 1605 |
+
"loss": 0.8055,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.8817017999807488,
|
| 1610 |
+
"grad_norm": 0.23171143233776093,
|
| 1611 |
+
"learning_rate": 3.643662883789878e-07,
|
| 1612 |
+
"loss": 0.8116,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.8855520261815382,
|
| 1617 |
+
"grad_norm": 0.2200126200914383,
|
| 1618 |
+
"learning_rate": 3.41358898202781e-07,
|
| 1619 |
+
"loss": 0.7873,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.8894022523823275,
|
| 1624 |
+
"grad_norm": 0.22081886231899261,
|
| 1625 |
+
"learning_rate": 3.190761058667741e-07,
|
| 1626 |
+
"loss": 0.7992,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.8932524785831167,
|
| 1631 |
+
"grad_norm": 0.22478991746902466,
|
| 1632 |
+
"learning_rate": 2.9752137720206463e-07,
|
| 1633 |
+
"loss": 0.807,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.8971027047839061,
|
| 1638 |
+
"grad_norm": 0.22948548197746277,
|
| 1639 |
+
"learning_rate": 2.7669806479785067e-07,
|
| 1640 |
+
"loss": 0.8105,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9009529309846953,
|
| 1645 |
+
"grad_norm": 0.23963341116905212,
|
| 1646 |
+
"learning_rate": 2.566094074799763e-07,
|
| 1647 |
+
"loss": 0.8059,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9048031571854847,
|
| 1652 |
+
"grad_norm": 0.22548210620880127,
|
| 1653 |
+
"learning_rate": 2.3725852980716557e-07,
|
| 1654 |
+
"loss": 0.7847,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.908653383386274,
|
| 1659 |
+
"grad_norm": 0.23088513314723969,
|
| 1660 |
+
"learning_rate": 2.1864844158503983e-07,
|
| 1661 |
+
"loss": 0.7986,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9125036095870632,
|
| 1666 |
+
"grad_norm": 0.25784873962402344,
|
| 1667 |
+
"learning_rate": 2.007820373979702e-07,
|
| 1668 |
+
"loss": 0.7907,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.9163538357878526,
|
| 1673 |
+
"grad_norm": 0.2319800704717636,
|
| 1674 |
+
"learning_rate": 1.8366209615886532e-07,
|
| 1675 |
+
"loss": 0.7995,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9202040619886418,
|
| 1680 |
+
"grad_norm": 0.2332865446805954,
|
| 1681 |
+
"learning_rate": 1.6729128067693613e-07,
|
| 1682 |
+
"loss": 0.8081,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.9240542881894311,
|
| 1687 |
+
"grad_norm": 0.24554863572120667,
|
| 1688 |
+
"learning_rate": 1.5167213724353426e-07,
|
| 1689 |
+
"loss": 0.8099,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9279045143902205,
|
| 1694 |
+
"grad_norm": 0.2250945270061493,
|
| 1695 |
+
"learning_rate": 1.3680709523610058e-07,
|
| 1696 |
+
"loss": 0.7926,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.9317547405910097,
|
| 1701 |
+
"grad_norm": 0.22248753905296326,
|
| 1702 |
+
"learning_rate": 1.2269846674030418e-07,
|
| 1703 |
+
"loss": 0.8154,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.935604966791799,
|
| 1708 |
+
"grad_norm": 0.22744449973106384,
|
| 1709 |
+
"learning_rate": 1.0934844619042595e-07,
|
| 1710 |
+
"loss": 0.8117,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.9394551929925883,
|
| 1715 |
+
"grad_norm": 0.2468804270029068,
|
| 1716 |
+
"learning_rate": 9.675911002803961e-08,
|
| 1717 |
+
"loss": 0.8018,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9433054191933776,
|
| 1722 |
+
"grad_norm": 0.24501512944698334,
|
| 1723 |
+
"learning_rate": 8.493241637904381e-08,
|
| 1724 |
+
"loss": 0.7985,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.9471556453941669,
|
| 1729 |
+
"grad_norm": 0.2378467172384262,
|
| 1730 |
+
"learning_rate": 7.387020474909689e-08,
|
| 1731 |
+
"loss": 0.8179,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9510058715949562,
|
| 1736 |
+
"grad_norm": 0.24185048043727875,
|
| 1737 |
+
"learning_rate": 6.357419573750856e-08,
|
| 1738 |
+
"loss": 0.814,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9548560977957455,
|
| 1743 |
+
"grad_norm": 0.24004718661308289,
|
| 1744 |
+
"learning_rate": 5.4045990769613365e-08,
|
| 1745 |
+
"loss": 0.796,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.9587063239965348,
|
| 1750 |
+
"grad_norm": 0.28036972880363464,
|
| 1751 |
+
"learning_rate": 4.5287071847694434e-08,
|
| 1752 |
+
"loss": 0.8013,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.9625565501973241,
|
| 1757 |
+
"grad_norm": 0.22754062712192535,
|
| 1758 |
+
"learning_rate": 3.7298801320467817e-08,
|
| 1759 |
+
"loss": 0.7943,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.9664067763981133,
|
| 1764 |
+
"grad_norm": 0.2318114936351776,
|
| 1765 |
+
"learning_rate": 3.0082421671192576e-08,
|
| 1766 |
+
"loss": 0.7873,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.9702570025989027,
|
| 1771 |
+
"grad_norm": 0.23515108227729797,
|
| 1772 |
+
"learning_rate": 2.3639055324410355e-08,
|
| 1773 |
+
"loss": 0.8095,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.974107228799692,
|
| 1778 |
+
"grad_norm": 0.2339082658290863,
|
| 1779 |
+
"learning_rate": 1.7969704471369477e-08,
|
| 1780 |
+
"loss": 0.7937,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.9779574550004813,
|
| 1785 |
+
"grad_norm": 0.22448225319385529,
|
| 1786 |
+
"learning_rate": 1.3075250914142413e-08,
|
| 1787 |
+
"loss": 0.7982,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.9818076812012706,
|
| 1792 |
+
"grad_norm": 0.22185440361499786,
|
| 1793 |
+
"learning_rate": 8.956455928476049e-09,
|
| 1794 |
+
"loss": 0.8079,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.9856579074020598,
|
| 1799 |
+
"grad_norm": 0.23270289599895477,
|
| 1800 |
+
"learning_rate": 5.6139601453786275e-09,
|
| 1801 |
+
"loss": 0.7995,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.9895081336028492,
|
| 1806 |
+
"grad_norm": 0.230127215385437,
|
| 1807 |
+
"learning_rate": 3.0482834514805602e-09,
|
| 1808 |
+
"loss": 0.7875,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.9933583598036385,
|
| 1813 |
+
"grad_norm": 0.21321165561676025,
|
| 1814 |
+
"learning_rate": 1.2598249081713366e-09,
|
| 1815 |
+
"loss": 0.7833,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.9972085860044277,
|
| 1820 |
+
"grad_norm": 0.2578992247581482,
|
| 1821 |
+
"learning_rate": 2.488626895302826e-10,
|
| 1822 |
+
"loss": 0.812,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.9999037443449803,
|
| 1827 |
+
"step": 2597,
|
| 1828 |
+
"total_flos": 3.777921305572016e+18,
|
| 1829 |
+
"train_loss": 0.8312876871745771,
|
| 1830 |
+
"train_runtime": 12239.4015,
|
| 1831 |
+
"train_samples_per_second": 54.325,
|
| 1832 |
+
"train_steps_per_second": 0.212
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 10,
|
| 1836 |
+
"max_steps": 2597,
|
| 1837 |
+
"num_input_tokens_seen": 418855352,
|
| 1838 |
+
"num_train_epochs": 1,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 3.777921305572016e+18,
|
| 1853 |
+
"train_batch_size": 8,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
continuous/qwen3/eupe_convnext_b/finetune/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c500d9f998ab0f7e541c6ee772fffa8d8dd30733852b9a6e766a6b94d2e318b
|
| 3 |
+
size 6264
|
continuous/qwen3/eupe_convnext_b/finetune/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
continuous/qwen3/eupe_convnext_b/pretrain/config.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_faster_video": false,
|
| 3 |
+
"add_time_instruction": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"Qwen3ForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 151645,
|
| 11 |
+
"faster_token_stride": 10,
|
| 12 |
+
"force_sample": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 2048,
|
| 16 |
+
"image_aspect_ratio": "square",
|
| 17 |
+
"image_crop_resolution": null,
|
| 18 |
+
"image_grid_pinpoints": null,
|
| 19 |
+
"image_split_resolution": null,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 6144,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention"
|
| 51 |
+
],
|
| 52 |
+
"max_position_embeddings": 40960,
|
| 53 |
+
"max_window_layers": 28,
|
| 54 |
+
"mm_hidden_size": 1024,
|
| 55 |
+
"mm_newline_position": "grid",
|
| 56 |
+
"mm_patch_merge_type": "flat",
|
| 57 |
+
"mm_projector_lr": null,
|
| 58 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 59 |
+
"mm_resampler_type": null,
|
| 60 |
+
"mm_spatial_pool_mode": "bilinear",
|
| 61 |
+
"mm_spatial_pool_stride": null,
|
| 62 |
+
"mm_tunable_parts": "mm_mlp_adapter",
|
| 63 |
+
"mm_use_im_patch_token": false,
|
| 64 |
+
"mm_use_im_start_end": false,
|
| 65 |
+
"mm_vision_select_feature": "patch",
|
| 66 |
+
"mm_vision_select_layer": -1,
|
| 67 |
+
"mm_vision_tower": "vtb_ssl:eupe",
|
| 68 |
+
"mm_vision_tower_lr": null,
|
| 69 |
+
"model_type": "qwen3",
|
| 70 |
+
"num_attention_heads": 16,
|
| 71 |
+
"num_hidden_layers": 28,
|
| 72 |
+
"num_key_value_heads": 8,
|
| 73 |
+
"pad_token_id": 151643,
|
| 74 |
+
"pos_skipping_range": 4096,
|
| 75 |
+
"rms_norm_eps": 1e-06,
|
| 76 |
+
"rope_scaling": null,
|
| 77 |
+
"rope_theta": 1000000,
|
| 78 |
+
"sliding_window": null,
|
| 79 |
+
"tie_word_embeddings": true,
|
| 80 |
+
"tokenizer_model_max_length": 8192,
|
| 81 |
+
"tokenizer_padding_side": "right",
|
| 82 |
+
"transformers_version": "4.57.6",
|
| 83 |
+
"use_cache": true,
|
| 84 |
+
"use_mm_proj": true,
|
| 85 |
+
"use_pos_skipping": false,
|
| 86 |
+
"use_sliding_window": false,
|
| 87 |
+
"vision_tower_pretrained": "/cache/ckpt/download/tokenizer/continuous/eupe/EUPE-ConvNeXt-B/EUPE-ConvNeXt-B.pt",
|
| 88 |
+
"vocab_size": 151936
|
| 89 |
+
}
|
continuous/qwen3/eupe_convnext_b/pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e2abd8a3ebbe2ad2c516b9315d30ab14371faee6c00b9e06ec7068f2ba33a36
|
| 3 |
+
size 25184312
|
continuous/qwen3/eupe_convnext_b/pretrain/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 545,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01834862385321101,
|
| 14 |
+
"grad_norm": 3.8056039810180664,
|
| 15 |
+
"learning_rate": 0.00032142857142857147,
|
| 16 |
+
"loss": 9.5653,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03669724770642202,
|
| 21 |
+
"grad_norm": 1.4165329933166504,
|
| 22 |
+
"learning_rate": 0.0006785714285714287,
|
| 23 |
+
"loss": 5.8209,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05504587155963303,
|
| 28 |
+
"grad_norm": 0.18496766686439514,
|
| 29 |
+
"learning_rate": 0.0009999907688176173,
|
| 30 |
+
"loss": 4.8789,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07339449541284404,
|
| 35 |
+
"grad_norm": 0.14828597009181976,
|
| 36 |
+
"learning_rate": 0.0009988834393115766,
|
| 37 |
+
"loss": 4.715,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09174311926605505,
|
| 42 |
+
"grad_norm": 0.09578010439872742,
|
| 43 |
+
"learning_rate": 0.0009959345572940772,
|
| 44 |
+
"loss": 4.6417,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11009174311926606,
|
| 49 |
+
"grad_norm": 0.09663919359445572,
|
| 50 |
+
"learning_rate": 0.0009911550081155983,
|
| 51 |
+
"loss": 4.5921,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.12844036697247707,
|
| 56 |
+
"grad_norm": 0.16856920719146729,
|
| 57 |
+
"learning_rate": 0.0009845624347566433,
|
| 58 |
+
"loss": 4.5384,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.14678899082568808,
|
| 63 |
+
"grad_norm": 0.13778077065944672,
|
| 64 |
+
"learning_rate": 0.0009761811727013547,
|
| 65 |
+
"loss": 4.505,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1651376146788991,
|
| 70 |
+
"grad_norm": 0.12277279049158096,
|
| 71 |
+
"learning_rate": 0.0009660421601067666,
|
| 72 |
+
"loss": 4.4563,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1834862385321101,
|
| 77 |
+
"grad_norm": 0.144491508603096,
|
| 78 |
+
"learning_rate": 0.0009541828235992926,
|
| 79 |
+
"loss": 4.4341,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.2018348623853211,
|
| 84 |
+
"grad_norm": 0.13881228864192963,
|
| 85 |
+
"learning_rate": 0.0009406469401200151,
|
| 86 |
+
"loss": 4.3998,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.22018348623853212,
|
| 91 |
+
"grad_norm": 0.13935136795043945,
|
| 92 |
+
"learning_rate": 0.0009254844753287492,
|
| 93 |
+
"loss": 4.3878,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23853211009174313,
|
| 98 |
+
"grad_norm": 0.13531967997550964,
|
| 99 |
+
"learning_rate": 0.0009087513991633924,
|
| 100 |
+
"loss": 4.3579,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.25688073394495414,
|
| 105 |
+
"grad_norm": 0.10001811385154724,
|
| 106 |
+
"learning_rate": 0.0008905094792353917,
|
| 107 |
+
"loss": 4.34,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.27522935779816515,
|
| 112 |
+
"grad_norm": 0.13013708591461182,
|
| 113 |
+
"learning_rate": 0.0008708260528239789,
|
| 114 |
+
"loss": 4.3409,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.29357798165137616,
|
| 119 |
+
"grad_norm": 0.10266247391700745,
|
| 120 |
+
"learning_rate": 0.0008497737783108237,
|
| 121 |
+
"loss": 4.3247,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.3119266055045872,
|
| 126 |
+
"grad_norm": 0.14916028082370758,
|
| 127 |
+
"learning_rate": 0.0008274303669726426,
|
| 128 |
+
"loss": 4.3102,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.3302752293577982,
|
| 133 |
+
"grad_norm": 0.1373901516199112,
|
| 134 |
+
"learning_rate": 0.0008038782961218137,
|
| 135 |
+
"loss": 4.2982,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.3486238532110092,
|
| 140 |
+
"grad_norm": 0.16083011031150818,
|
| 141 |
+
"learning_rate": 0.000779204504653894,
|
| 142 |
+
"loss": 4.2968,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3669724770642202,
|
| 147 |
+
"grad_norm": 0.13746394217014313,
|
| 148 |
+
"learning_rate": 0.0007535000721258791,
|
| 149 |
+
"loss": 4.2766,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3853211009174312,
|
| 154 |
+
"grad_norm": 0.13349492847919464,
|
| 155 |
+
"learning_rate": 0.00072685988254984,
|
| 156 |
+
"loss": 4.2688,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4036697247706422,
|
| 161 |
+
"grad_norm": 0.1190626248717308,
|
| 162 |
+
"learning_rate": 0.0006993822741429906,
|
| 163 |
+
"loss": 4.2828,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.42201834862385323,
|
| 168 |
+
"grad_norm": 0.11626529693603516,
|
| 169 |
+
"learning_rate": 0.0006711686763270818,
|
| 170 |
+
"loss": 4.2775,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.44036697247706424,
|
| 175 |
+
"grad_norm": 0.10219736397266388,
|
| 176 |
+
"learning_rate": 0.0006423232353170831,
|
| 177 |
+
"loss": 4.2655,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.45871559633027525,
|
| 182 |
+
"grad_norm": 0.12429136037826538,
|
| 183 |
+
"learning_rate": 0.0006129524296812335,
|
| 184 |
+
"loss": 4.2462,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.47706422018348627,
|
| 189 |
+
"grad_norm": 0.10710363835096359,
|
| 190 |
+
"learning_rate": 0.0005831646772915651,
|
| 191 |
+
"loss": 4.26,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.4954128440366973,
|
| 196 |
+
"grad_norm": 0.11219484359025955,
|
| 197 |
+
"learning_rate": 0.0005530699351157809,
|
| 198 |
+
"loss": 4.2417,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.5137614678899083,
|
| 203 |
+
"grad_norm": 0.11874975264072418,
|
| 204 |
+
"learning_rate": 0.0005227792933277943,
|
| 205 |
+
"loss": 4.2395,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.5321100917431193,
|
| 210 |
+
"grad_norm": 0.11727134138345718,
|
| 211 |
+
"learning_rate": 0.0004924045652352092,
|
| 212 |
+
"loss": 4.2253,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.5504587155963303,
|
| 217 |
+
"grad_norm": 0.11342450976371765,
|
| 218 |
+
"learning_rate": 0.0004620578745374633,
|
| 219 |
+
"loss": 4.227,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.5688073394495413,
|
| 224 |
+
"grad_norm": 0.13225379586219788,
|
| 225 |
+
"learning_rate": 0.0004318512414382059,
|
| 226 |
+
"loss": 4.2426,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.5871559633027523,
|
| 231 |
+
"grad_norm": 0.10023616999387741,
|
| 232 |
+
"learning_rate": 0.0004018961691397148,
|
| 233 |
+
"loss": 4.2304,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6055045871559633,
|
| 238 |
+
"grad_norm": 0.11039053648710251,
|
| 239 |
+
"learning_rate": 0.0003723032322457458,
|
| 240 |
+
"loss": 4.2131,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.6238532110091743,
|
| 245 |
+
"grad_norm": 0.11465641856193542,
|
| 246 |
+
"learning_rate": 0.0003431816685921625,
|
| 247 |
+
"loss": 4.2151,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.6422018348623854,
|
| 252 |
+
"grad_norm": 0.09876559674739838,
|
| 253 |
+
"learning_rate": 0.0003146389760120416,
|
| 254 |
+
"loss": 4.2198,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.6605504587155964,
|
| 259 |
+
"grad_norm": 0.10564030706882477,
|
| 260 |
+
"learning_rate": 0.00028678051552373485,
|
| 261 |
+
"loss": 4.2263,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.6788990825688074,
|
| 266 |
+
"grad_norm": 0.1000557690858841,
|
| 267 |
+
"learning_rate": 0.0002597091224066581,
|
| 268 |
+
"loss": 4.2104,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.6972477064220184,
|
| 273 |
+
"grad_norm": 0.10158879309892654,
|
| 274 |
+
"learning_rate": 0.0002335247266004629,
|
| 275 |
+
"loss": 4.2272,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.7155963302752294,
|
| 280 |
+
"grad_norm": 0.10967017710208893,
|
| 281 |
+
"learning_rate": 0.000208323983828828,
|
| 282 |
+
"loss": 4.2076,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.7339449541284404,
|
| 287 |
+
"grad_norm": 0.1044796034693718,
|
| 288 |
+
"learning_rate": 0.0001841999188095224,
|
| 289 |
+
"loss": 4.2112,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.7522935779816514,
|
| 294 |
+
"grad_norm": 0.09305652976036072,
|
| 295 |
+
"learning_rate": 0.00016124158186777676,
|
| 296 |
+
"loss": 4.21,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.7706422018348624,
|
| 301 |
+
"grad_norm": 0.112674281001091,
|
| 302 |
+
"learning_rate": 0.0001395337202205211,
|
| 303 |
+
"loss": 4.2013,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.7889908256880734,
|
| 308 |
+
"grad_norm": 0.08949209749698639,
|
| 309 |
+
"learning_rate": 0.00011915646514489292,
|
| 310 |
+
"loss": 4.2089,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.8073394495412844,
|
| 315 |
+
"grad_norm": 0.11475195735692978,
|
| 316 |
+
"learning_rate": 0.00010018503618578818,
|
| 317 |
+
"loss": 4.2155,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.8256880733944955,
|
| 322 |
+
"grad_norm": 0.09541667997837067,
|
| 323 |
+
"learning_rate": 8.268946349432583e-05,
|
| 324 |
+
"loss": 4.2074,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.8440366972477065,
|
| 329 |
+
"grad_norm": 0.09782084077596664,
|
| 330 |
+
"learning_rate": 6.673432932217117e-05,
|
| 331 |
+
"loss": 4.1983,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.8623853211009175,
|
| 336 |
+
"grad_norm": 0.09632977843284607,
|
| 337 |
+
"learning_rate": 5.2378529625954684e-05,
|
| 338 |
+
"loss": 4.1878,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.8807339449541285,
|
| 343 |
+
"grad_norm": 0.09285271912813187,
|
| 344 |
+
"learning_rate": 3.9675056661785556e-05,
|
| 345 |
+
"loss": 4.196,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.8990825688073395,
|
| 350 |
+
"grad_norm": 0.09220317751169205,
|
| 351 |
+
"learning_rate": 2.8670803372379538e-05,
|
| 352 |
+
"loss": 4.1974,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.9174311926605505,
|
| 357 |
+
"grad_norm": 0.08745542913675308,
|
| 358 |
+
"learning_rate": 1.9406390288876586e-05,
|
| 359 |
+
"loss": 4.2114,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.9357798165137615,
|
| 364 |
+
"grad_norm": 0.08986830711364746,
|
| 365 |
+
"learning_rate": 1.191601558631461e-05,
|
| 366 |
+
"loss": 4.2002,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.9541284403669725,
|
| 371 |
+
"grad_norm": 0.08838362246751785,
|
| 372 |
+
"learning_rate": 6.2273288462559305e-06,
|
| 373 |
+
"loss": 4.1896,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.9724770642201835,
|
| 378 |
+
"grad_norm": 0.08508139103651047,
|
| 379 |
+
"learning_rate": 2.361328992552314e-06,
|
| 380 |
+
"loss": 4.1948,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.9908256880733946,
|
| 385 |
+
"grad_norm": 0.09262628108263016,
|
| 386 |
+
"learning_rate": 3.322867770044202e-07,
|
| 387 |
+
"loss": 4.1985,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0,
|
| 392 |
+
"step": 545,
|
| 393 |
+
"total_flos": 2.989368819014697e+17,
|
| 394 |
+
"train_loss": 4.424892667017946,
|
| 395 |
+
"train_runtime": 5088.946,
|
| 396 |
+
"train_samples_per_second": 109.675,
|
| 397 |
+
"train_steps_per_second": 0.107
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"logging_steps": 10,
|
| 401 |
+
"max_steps": 545,
|
| 402 |
+
"num_input_tokens_seen": 33142912,
|
| 403 |
+
"num_train_epochs": 1,
|
| 404 |
+
"save_steps": 500,
|
| 405 |
+
"stateful_callbacks": {
|
| 406 |
+
"TrainerControl": {
|
| 407 |
+
"args": {
|
| 408 |
+
"should_epoch_stop": false,
|
| 409 |
+
"should_evaluate": false,
|
| 410 |
+
"should_log": false,
|
| 411 |
+
"should_save": false,
|
| 412 |
+
"should_training_stop": false
|
| 413 |
+
},
|
| 414 |
+
"attributes": {}
|
| 415 |
+
}
|
| 416 |
+
},
|
| 417 |
+
"total_flos": 2.989368819014697e+17,
|
| 418 |
+
"train_batch_size": 64,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
continuous/qwen3/eupe_vit_b/finetune/added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
continuous/qwen3/eupe_vit_b/finetune/chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|