Upload folder using huggingface_hub
Browse files- config.json +64 -0
- experiment_cfg/metadata.json +363 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
- trainer_state.json +743 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"action_dim": 32,
|
| 3 |
+
"action_head_cfg": {
|
| 4 |
+
"action_dim": 32,
|
| 5 |
+
"action_horizon": 16,
|
| 6 |
+
"add_pos_embed": true,
|
| 7 |
+
"backbone_embedding_dim": 2048,
|
| 8 |
+
"diffusion_model_cfg": {
|
| 9 |
+
"attention_head_dim": 48,
|
| 10 |
+
"cross_attention_dim": 2048,
|
| 11 |
+
"dropout": 0.2,
|
| 12 |
+
"final_dropout": true,
|
| 13 |
+
"interleave_self_attention": true,
|
| 14 |
+
"norm_type": "ada_norm",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_layers": 16,
|
| 17 |
+
"output_dim": 1024,
|
| 18 |
+
"positional_embeddings": null
|
| 19 |
+
},
|
| 20 |
+
"hidden_size": 1024,
|
| 21 |
+
"input_embedding_dim": 1536,
|
| 22 |
+
"max_action_dim": 32,
|
| 23 |
+
"max_state_dim": 64,
|
| 24 |
+
"model_dtype": "float32",
|
| 25 |
+
"noise_beta_alpha": 1.5,
|
| 26 |
+
"noise_beta_beta": 1.0,
|
| 27 |
+
"noise_s": 0.999,
|
| 28 |
+
"num_inference_timesteps": 4,
|
| 29 |
+
"num_target_vision_tokens": 32,
|
| 30 |
+
"num_timestep_buckets": 1000,
|
| 31 |
+
"tune_diffusion_model": true,
|
| 32 |
+
"tune_projector": true,
|
| 33 |
+
"use_vlln": true,
|
| 34 |
+
"vl_self_attention_cfg": {
|
| 35 |
+
"attention_head_dim": 64,
|
| 36 |
+
"dropout": 0.2,
|
| 37 |
+
"final_dropout": true,
|
| 38 |
+
"num_attention_heads": 32,
|
| 39 |
+
"num_layers": 4,
|
| 40 |
+
"positional_embeddings": null
|
| 41 |
+
}
|
| 42 |
+
},
|
| 43 |
+
"action_horizon": 16,
|
| 44 |
+
"architectures": [
|
| 45 |
+
"GR00T_N1_5"
|
| 46 |
+
],
|
| 47 |
+
"attn_implementation": null,
|
| 48 |
+
"backbone_cfg": {
|
| 49 |
+
"eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
|
| 50 |
+
"load_bf16": false,
|
| 51 |
+
"project_to_dim": null,
|
| 52 |
+
"reproject_vision": false,
|
| 53 |
+
"select_layer": 12,
|
| 54 |
+
"tune_llm": false,
|
| 55 |
+
"tune_visual": true,
|
| 56 |
+
"use_flash_attention": true
|
| 57 |
+
},
|
| 58 |
+
"compute_dtype": "bfloat16",
|
| 59 |
+
"hidden_size": 2048,
|
| 60 |
+
"model_dtype": "float32",
|
| 61 |
+
"model_type": "gr00t_n1_5",
|
| 62 |
+
"torch_dtype": "bfloat16",
|
| 63 |
+
"transformers_version": "4.51.3"
|
| 64 |
+
}
|
experiment_cfg/metadata.json
ADDED
|
@@ -0,0 +1,363 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"new_embodiment": {
|
| 3 |
+
"statistics": {
|
| 4 |
+
"state": {
|
| 5 |
+
"left_arm": {
|
| 6 |
+
"max": [
|
| 7 |
+
31.097339630126953,
|
| 8 |
+
87.07456970214844,
|
| 9 |
+
99.29947662353516,
|
| 10 |
+
75.39047241210938,
|
| 11 |
+
-39.19413757324219
|
| 12 |
+
],
|
| 13 |
+
"min": [
|
| 14 |
+
-37.65949630737305,
|
| 15 |
+
-96.6693115234375,
|
| 16 |
+
-98.07355499267578,
|
| 17 |
+
-72.19047546386719,
|
| 18 |
+
-59.511600494384766
|
| 19 |
+
],
|
| 20 |
+
"mean": [
|
| 21 |
+
5.31805419921875,
|
| 22 |
+
-73.24604797363281,
|
| 23 |
+
84.5744857788086,
|
| 24 |
+
53.659244537353516,
|
| 25 |
+
-52.5255012512207
|
| 26 |
+
],
|
| 27 |
+
"std": [
|
| 28 |
+
8.872116088867188,
|
| 29 |
+
47.90742492675781,
|
| 30 |
+
32.04811096191406,
|
| 31 |
+
43.50612258911133,
|
| 32 |
+
3.739243984222412
|
| 33 |
+
],
|
| 34 |
+
"q01": [
|
| 35 |
+
-24.316442489624023,
|
| 36 |
+
-96.6693115234375,
|
| 37 |
+
-18.388792037963867,
|
| 38 |
+
-54.05714416503906,
|
| 39 |
+
-56.58119583129883
|
| 40 |
+
],
|
| 41 |
+
"q99": [
|
| 42 |
+
24.826831817626953,
|
| 43 |
+
49.59846878051758,
|
| 44 |
+
99.2122573852539,
|
| 45 |
+
75.29190063476562,
|
| 46 |
+
-40.17094039916992
|
| 47 |
+
]
|
| 48 |
+
},
|
| 49 |
+
"gripper1": {
|
| 50 |
+
"max": [
|
| 51 |
+
65.40178680419922
|
| 52 |
+
],
|
| 53 |
+
"min": [
|
| 54 |
+
0.1488095223903656
|
| 55 |
+
],
|
| 56 |
+
"mean": [
|
| 57 |
+
14.567317008972168
|
| 58 |
+
],
|
| 59 |
+
"std": [
|
| 60 |
+
7.672760486602783
|
| 61 |
+
],
|
| 62 |
+
"q01": [
|
| 63 |
+
0.5208333134651184
|
| 64 |
+
],
|
| 65 |
+
"q99": [
|
| 66 |
+
44.12202453613281
|
| 67 |
+
]
|
| 68 |
+
},
|
| 69 |
+
"right_arm": {
|
| 70 |
+
"max": [
|
| 71 |
+
39.934234619140625,
|
| 72 |
+
76.15731048583984,
|
| 73 |
+
99.3824462890625,
|
| 74 |
+
76.23355102539062,
|
| 75 |
+
38.59306335449219
|
| 76 |
+
],
|
| 77 |
+
"min": [
|
| 78 |
+
-33.64998245239258,
|
| 79 |
+
-98.0335922241211,
|
| 80 |
+
-49.80149841308594,
|
| 81 |
+
-95.80592346191406,
|
| 82 |
+
-64.38690948486328
|
| 83 |
+
],
|
| 84 |
+
"mean": [
|
| 85 |
+
1.84208345413208,
|
| 86 |
+
-20.025339126586914,
|
| 87 |
+
60.03025817871094,
|
| 88 |
+
-8.195182800292969,
|
| 89 |
+
-47.5246467590332
|
| 90 |
+
],
|
| 91 |
+
"std": [
|
| 92 |
+
11.903837203979492,
|
| 93 |
+
62.466819763183594,
|
| 94 |
+
35.845306396484375,
|
| 95 |
+
63.938392639160156,
|
| 96 |
+
17.236434936523438
|
| 97 |
+
],
|
| 98 |
+
"q01": [
|
| 99 |
+
-22.908294677734375,
|
| 100 |
+
-97.6239242553711,
|
| 101 |
+
-20.688133239746094,
|
| 102 |
+
-94.90131378173828,
|
| 103 |
+
-58.133853912353516
|
| 104 |
+
],
|
| 105 |
+
"q99": [
|
| 106 |
+
29.557910919189453,
|
| 107 |
+
67.06267547607422,
|
| 108 |
+
99.20600128173828,
|
| 109 |
+
75.9868392944336,
|
| 110 |
+
22.66731834411621
|
| 111 |
+
]
|
| 112 |
+
},
|
| 113 |
+
"gripper2": {
|
| 114 |
+
"max": [
|
| 115 |
+
91.74726104736328
|
| 116 |
+
],
|
| 117 |
+
"min": [
|
| 118 |
+
2.3210830688476562
|
| 119 |
+
],
|
| 120 |
+
"mean": [
|
| 121 |
+
11.800792694091797
|
| 122 |
+
],
|
| 123 |
+
"std": [
|
| 124 |
+
16.004310607910156
|
| 125 |
+
],
|
| 126 |
+
"q01": [
|
| 127 |
+
2.3855576515197754
|
| 128 |
+
],
|
| 129 |
+
"q99": [
|
| 130 |
+
67.69825744628906
|
| 131 |
+
]
|
| 132 |
+
}
|
| 133 |
+
},
|
| 134 |
+
"action": {
|
| 135 |
+
"left_arm": {
|
| 136 |
+
"max": [
|
| 137 |
+
33.853736877441406,
|
| 138 |
+
90.75993347167969,
|
| 139 |
+
100.0,
|
| 140 |
+
76.46582794189453,
|
| 141 |
+
-39.19413757324219
|
| 142 |
+
],
|
| 143 |
+
"min": [
|
| 144 |
+
-37.88003158569336,
|
| 145 |
+
-96.45941162109375,
|
| 146 |
+
-99.63386535644531,
|
| 147 |
+
-73.87788391113281,
|
| 148 |
+
-59.85348129272461
|
| 149 |
+
],
|
| 150 |
+
"mean": [
|
| 151 |
+
5.2239155769348145,
|
| 152 |
+
-73.25548553466797,
|
| 153 |
+
84.9612808227539,
|
| 154 |
+
54.047916412353516,
|
| 155 |
+
-52.520851135253906
|
| 156 |
+
],
|
| 157 |
+
"std": [
|
| 158 |
+
8.930574417114258,
|
| 159 |
+
47.422481536865234,
|
| 160 |
+
32.83000946044922,
|
| 161 |
+
44.03689193725586,
|
| 162 |
+
3.7615270614624023
|
| 163 |
+
],
|
| 164 |
+
"q01": [
|
| 165 |
+
-24.23993492126465,
|
| 166 |
+
-96.37305450439453,
|
| 167 |
+
-19.450801849365234,
|
| 168 |
+
-54.6300048828125,
|
| 169 |
+
-56.630035400390625
|
| 170 |
+
],
|
| 171 |
+
"q99": [
|
| 172 |
+
25.390304565429688,
|
| 173 |
+
48.96372985839844,
|
| 174 |
+
99.81693267822266,
|
| 175 |
+
76.142333984375,
|
| 176 |
+
-40.17094039916992
|
| 177 |
+
]
|
| 178 |
+
},
|
| 179 |
+
"gripper1": {
|
| 180 |
+
"max": [
|
| 181 |
+
65.94789123535156
|
| 182 |
+
],
|
| 183 |
+
"min": [
|
| 184 |
+
0.0
|
| 185 |
+
],
|
| 186 |
+
"mean": [
|
| 187 |
+
14.2286376953125
|
| 188 |
+
],
|
| 189 |
+
"std": [
|
| 190 |
+
8.476670265197754
|
| 191 |
+
],
|
| 192 |
+
"q01": [
|
| 193 |
+
0.0
|
| 194 |
+
],
|
| 195 |
+
"q99": [
|
| 196 |
+
44.290208244323466
|
| 197 |
+
]
|
| 198 |
+
},
|
| 199 |
+
"right_arm": {
|
| 200 |
+
"max": [
|
| 201 |
+
40.137088775634766,
|
| 202 |
+
75.02074432373047,
|
| 203 |
+
99.91039276123047,
|
| 204 |
+
78.80850982666016,
|
| 205 |
+
39.21210479736328
|
| 206 |
+
],
|
| 207 |
+
"min": [
|
| 208 |
+
-33.43488311767578,
|
| 209 |
+
-99.004150390625,
|
| 210 |
+
-51.7921142578125,
|
| 211 |
+
-96.85106658935547,
|
| 212 |
+
-64.67518615722656
|
| 213 |
+
],
|
| 214 |
+
"mean": [
|
| 215 |
+
1.85464346408844,
|
| 216 |
+
-21.57091522216797,
|
| 217 |
+
59.33572769165039,
|
| 218 |
+
-8.245699882507324,
|
| 219 |
+
-47.39208221435547
|
| 220 |
+
],
|
| 221 |
+
"std": [
|
| 222 |
+
11.910544395446777,
|
| 223 |
+
61.55514907836914,
|
| 224 |
+
36.57780838012695,
|
| 225 |
+
64.52499389648438,
|
| 226 |
+
17.27439308166504
|
| 227 |
+
],
|
| 228 |
+
"q01": [
|
| 229 |
+
-22.848438262939453,
|
| 230 |
+
-97.84232330322266,
|
| 231 |
+
-22.307348251342766,
|
| 232 |
+
-95.82978820800781,
|
| 233 |
+
-58.20506286621094
|
| 234 |
+
],
|
| 235 |
+
"q99": [
|
| 236 |
+
29.702970504760742,
|
| 237 |
+
65.3941879272461,
|
| 238 |
+
99.46236419677734,
|
| 239 |
+
77.7872314453125,
|
| 240 |
+
22.82807159423828
|
| 241 |
+
]
|
| 242 |
+
},
|
| 243 |
+
"gripper2": {
|
| 244 |
+
"max": [
|
| 245 |
+
92.38652801513672
|
| 246 |
+
],
|
| 247 |
+
"min": [
|
| 248 |
+
0.14641287922859192
|
| 249 |
+
],
|
| 250 |
+
"mean": [
|
| 251 |
+
9.158825874328613
|
| 252 |
+
],
|
| 253 |
+
"std": [
|
| 254 |
+
17.096576690673828
|
| 255 |
+
],
|
| 256 |
+
"q01": [
|
| 257 |
+
1.6105417013168335
|
| 258 |
+
],
|
| 259 |
+
"q99": [
|
| 260 |
+
68.52123260498047
|
| 261 |
+
]
|
| 262 |
+
}
|
| 263 |
+
}
|
| 264 |
+
},
|
| 265 |
+
"modalities": {
|
| 266 |
+
"video": {
|
| 267 |
+
"right": {
|
| 268 |
+
"resolution": [
|
| 269 |
+
640,
|
| 270 |
+
480
|
| 271 |
+
],
|
| 272 |
+
"channels": 3,
|
| 273 |
+
"fps": 30.0
|
| 274 |
+
},
|
| 275 |
+
"left": {
|
| 276 |
+
"resolution": [
|
| 277 |
+
640,
|
| 278 |
+
480
|
| 279 |
+
],
|
| 280 |
+
"channels": 3,
|
| 281 |
+
"fps": 30.0
|
| 282 |
+
},
|
| 283 |
+
"top_rgb": {
|
| 284 |
+
"resolution": [
|
| 285 |
+
640,
|
| 286 |
+
480
|
| 287 |
+
],
|
| 288 |
+
"channels": 3,
|
| 289 |
+
"fps": 30.0
|
| 290 |
+
}
|
| 291 |
+
},
|
| 292 |
+
"state": {
|
| 293 |
+
"left_arm": {
|
| 294 |
+
"absolute": true,
|
| 295 |
+
"rotation_type": null,
|
| 296 |
+
"shape": [
|
| 297 |
+
5
|
| 298 |
+
],
|
| 299 |
+
"continuous": true
|
| 300 |
+
},
|
| 301 |
+
"gripper1": {
|
| 302 |
+
"absolute": true,
|
| 303 |
+
"rotation_type": null,
|
| 304 |
+
"shape": [
|
| 305 |
+
1
|
| 306 |
+
],
|
| 307 |
+
"continuous": true
|
| 308 |
+
},
|
| 309 |
+
"right_arm": {
|
| 310 |
+
"absolute": true,
|
| 311 |
+
"rotation_type": null,
|
| 312 |
+
"shape": [
|
| 313 |
+
5
|
| 314 |
+
],
|
| 315 |
+
"continuous": true
|
| 316 |
+
},
|
| 317 |
+
"gripper2": {
|
| 318 |
+
"absolute": true,
|
| 319 |
+
"rotation_type": null,
|
| 320 |
+
"shape": [
|
| 321 |
+
1
|
| 322 |
+
],
|
| 323 |
+
"continuous": true
|
| 324 |
+
}
|
| 325 |
+
},
|
| 326 |
+
"action": {
|
| 327 |
+
"left_arm": {
|
| 328 |
+
"absolute": true,
|
| 329 |
+
"rotation_type": null,
|
| 330 |
+
"shape": [
|
| 331 |
+
5
|
| 332 |
+
],
|
| 333 |
+
"continuous": true
|
| 334 |
+
},
|
| 335 |
+
"gripper1": {
|
| 336 |
+
"absolute": true,
|
| 337 |
+
"rotation_type": null,
|
| 338 |
+
"shape": [
|
| 339 |
+
1
|
| 340 |
+
],
|
| 341 |
+
"continuous": true
|
| 342 |
+
},
|
| 343 |
+
"right_arm": {
|
| 344 |
+
"absolute": true,
|
| 345 |
+
"rotation_type": null,
|
| 346 |
+
"shape": [
|
| 347 |
+
5
|
| 348 |
+
],
|
| 349 |
+
"continuous": true
|
| 350 |
+
},
|
| 351 |
+
"gripper2": {
|
| 352 |
+
"absolute": true,
|
| 353 |
+
"rotation_type": null,
|
| 354 |
+
"shape": [
|
| 355 |
+
1
|
| 356 |
+
],
|
| 357 |
+
"continuous": true
|
| 358 |
+
}
|
| 359 |
+
}
|
| 360 |
+
},
|
| 361 |
+
"embodiment_tag": "new_embodiment"
|
| 362 |
+
}
|
| 363 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ad0600328b30fafe76a295afa31ea8d28edba7df32408932d422d6e9fca2a26
|
| 3 |
+
size 4999367032
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e508f2c70e9f0c50a51672c43b3ede7ee8e1afe0a4545cc891bc2e7a402bfd39
|
| 3 |
+
size 2586705312
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
trainer_state.json
ADDED
|
@@ -0,0 +1,743 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.1520737327188941,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.01152073732718894,
|
| 14 |
+
"grad_norm": 1.6566706895828247,
|
| 15 |
+
"learning_rate": 1.8e-05,
|
| 16 |
+
"loss": 0.9518,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.02304147465437788,
|
| 21 |
+
"grad_norm": 0.8684583902359009,
|
| 22 |
+
"learning_rate": 3.8e-05,
|
| 23 |
+
"loss": 0.1942,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03456221198156682,
|
| 28 |
+
"grad_norm": 0.6393296718597412,
|
| 29 |
+
"learning_rate": 5.8e-05,
|
| 30 |
+
"loss": 0.125,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.04608294930875576,
|
| 35 |
+
"grad_norm": 0.35887211561203003,
|
| 36 |
+
"learning_rate": 7.800000000000001e-05,
|
| 37 |
+
"loss": 0.0976,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0576036866359447,
|
| 42 |
+
"grad_norm": 0.7574331164360046,
|
| 43 |
+
"learning_rate": 9.8e-05,
|
| 44 |
+
"loss": 0.0844,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06912442396313365,
|
| 49 |
+
"grad_norm": 0.7963987588882446,
|
| 50 |
+
"learning_rate": 9.997785653888835e-05,
|
| 51 |
+
"loss": 0.0697,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.08064516129032258,
|
| 56 |
+
"grad_norm": 0.7024580240249634,
|
| 57 |
+
"learning_rate": 9.990133642141359e-05,
|
| 58 |
+
"loss": 0.0661,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.09216589861751152,
|
| 63 |
+
"grad_norm": 0.6337135434150696,
|
| 64 |
+
"learning_rate": 9.977024992520602e-05,
|
| 65 |
+
"loss": 0.0577,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.10368663594470046,
|
| 70 |
+
"grad_norm": 0.3510747253894806,
|
| 71 |
+
"learning_rate": 9.95847403914247e-05,
|
| 72 |
+
"loss": 0.0537,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1152073732718894,
|
| 77 |
+
"grad_norm": 0.4563221037387848,
|
| 78 |
+
"learning_rate": 9.934501067202117e-05,
|
| 79 |
+
"loss": 0.052,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.12672811059907835,
|
| 84 |
+
"grad_norm": 0.3978365957736969,
|
| 85 |
+
"learning_rate": 9.905132290792394e-05,
|
| 86 |
+
"loss": 0.0445,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.1382488479262673,
|
| 91 |
+
"grad_norm": 0.40588557720184326,
|
| 92 |
+
"learning_rate": 9.870399824239117e-05,
|
| 93 |
+
"loss": 0.0427,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1497695852534562,
|
| 98 |
+
"grad_norm": 0.5251225233078003,
|
| 99 |
+
"learning_rate": 9.830341646984521e-05,
|
| 100 |
+
"loss": 0.0378,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.16129032258064516,
|
| 105 |
+
"grad_norm": 0.38299882411956787,
|
| 106 |
+
"learning_rate": 9.785001562057309e-05,
|
| 107 |
+
"loss": 0.0404,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1728110599078341,
|
| 112 |
+
"grad_norm": 0.2926655411720276,
|
| 113 |
+
"learning_rate": 9.734429148174675e-05,
|
| 114 |
+
"loss": 0.0395,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.18433179723502305,
|
| 119 |
+
"grad_norm": 0.4306221306324005,
|
| 120 |
+
"learning_rate": 9.6786797055287e-05,
|
| 121 |
+
"loss": 0.0361,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.195852534562212,
|
| 126 |
+
"grad_norm": 0.3377871811389923,
|
| 127 |
+
"learning_rate": 9.617814195316411e-05,
|
| 128 |
+
"loss": 0.035,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.2073732718894009,
|
| 133 |
+
"grad_norm": 0.40714597702026367,
|
| 134 |
+
"learning_rate": 9.551899173079607e-05,
|
| 135 |
+
"loss": 0.0332,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.21889400921658986,
|
| 140 |
+
"grad_norm": 0.40230241417884827,
|
| 141 |
+
"learning_rate": 9.481006715927351e-05,
|
| 142 |
+
"loss": 0.0321,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.2304147465437788,
|
| 147 |
+
"grad_norm": 0.41335731744766235,
|
| 148 |
+
"learning_rate": 9.405214343720707e-05,
|
| 149 |
+
"loss": 0.0319,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.24193548387096775,
|
| 154 |
+
"grad_norm": 0.20366613566875458,
|
| 155 |
+
"learning_rate": 9.32460493430591e-05,
|
| 156 |
+
"loss": 0.0276,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2534562211981567,
|
| 161 |
+
"grad_norm": 0.24148087203502655,
|
| 162 |
+
"learning_rate": 9.239266632888659e-05,
|
| 163 |
+
"loss": 0.0331,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.26497695852534564,
|
| 168 |
+
"grad_norm": 0.31080925464630127,
|
| 169 |
+
"learning_rate": 9.14929275564863e-05,
|
| 170 |
+
"loss": 0.0305,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2764976958525346,
|
| 175 |
+
"grad_norm": 0.2609555125236511,
|
| 176 |
+
"learning_rate": 9.0547816876996e-05,
|
| 177 |
+
"loss": 0.0277,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2880184331797235,
|
| 182 |
+
"grad_norm": 0.35045984387397766,
|
| 183 |
+
"learning_rate": 8.955836775506776e-05,
|
| 184 |
+
"loss": 0.0254,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2995391705069124,
|
| 189 |
+
"grad_norm": 0.2834389805793762,
|
| 190 |
+
"learning_rate": 8.852566213878947e-05,
|
| 191 |
+
"loss": 0.0272,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.31105990783410137,
|
| 196 |
+
"grad_norm": 0.48804351687431335,
|
| 197 |
+
"learning_rate": 8.745082927659047e-05,
|
| 198 |
+
"loss": 0.0273,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3225806451612903,
|
| 203 |
+
"grad_norm": 0.36219653487205505,
|
| 204 |
+
"learning_rate": 8.633504448242505e-05,
|
| 205 |
+
"loss": 0.0274,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.33410138248847926,
|
| 210 |
+
"grad_norm": 0.28334954380989075,
|
| 211 |
+
"learning_rate": 8.517952785058385e-05,
|
| 212 |
+
"loss": 0.0244,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.3456221198156682,
|
| 217 |
+
"grad_norm": 0.37658995389938354,
|
| 218 |
+
"learning_rate": 8.398554292153866e-05,
|
| 219 |
+
"loss": 0.0234,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.35714285714285715,
|
| 224 |
+
"grad_norm": 0.4306541979312897,
|
| 225 |
+
"learning_rate": 8.275439530027948e-05,
|
| 226 |
+
"loss": 0.0255,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.3686635944700461,
|
| 231 |
+
"grad_norm": 0.3718424141407013,
|
| 232 |
+
"learning_rate": 8.148743122865463e-05,
|
| 233 |
+
"loss": 0.026,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.38018433179723504,
|
| 238 |
+
"grad_norm": 0.40250667929649353,
|
| 239 |
+
"learning_rate": 8.018603611327504e-05,
|
| 240 |
+
"loss": 0.0234,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.391705069124424,
|
| 245 |
+
"grad_norm": 0.3549322187900543,
|
| 246 |
+
"learning_rate": 7.88516330105925e-05,
|
| 247 |
+
"loss": 0.0246,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.4032258064516129,
|
| 252 |
+
"grad_norm": 0.3418162763118744,
|
| 253 |
+
"learning_rate": 7.748568107080832e-05,
|
| 254 |
+
"loss": 0.0214,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.4147465437788018,
|
| 259 |
+
"grad_norm": 0.4048205614089966,
|
| 260 |
+
"learning_rate": 7.608967394231387e-05,
|
| 261 |
+
"loss": 0.021,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.42626728110599077,
|
| 266 |
+
"grad_norm": 0.2962040603160858,
|
| 267 |
+
"learning_rate": 7.466513813840825e-05,
|
| 268 |
+
"loss": 0.0229,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4377880184331797,
|
| 273 |
+
"grad_norm": 0.26386022567749023,
|
| 274 |
+
"learning_rate": 7.32136313680782e-05,
|
| 275 |
+
"loss": 0.021,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.44930875576036866,
|
| 280 |
+
"grad_norm": 0.18067125976085663,
|
| 281 |
+
"learning_rate": 7.173674083266624e-05,
|
| 282 |
+
"loss": 0.0207,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.4608294930875576,
|
| 287 |
+
"grad_norm": 0.24752575159072876,
|
| 288 |
+
"learning_rate": 7.023608149028937e-05,
|
| 289 |
+
"loss": 0.023,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.47235023041474655,
|
| 294 |
+
"grad_norm": 0.24577368795871735,
|
| 295 |
+
"learning_rate": 6.871329428990602e-05,
|
| 296 |
+
"loss": 0.0199,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4838709677419355,
|
| 301 |
+
"grad_norm": 0.20847314596176147,
|
| 302 |
+
"learning_rate": 6.71700443769625e-05,
|
| 303 |
+
"loss": 0.0214,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.49539170506912444,
|
| 308 |
+
"grad_norm": 0.19042600691318512,
|
| 309 |
+
"learning_rate": 6.56080192725808e-05,
|
| 310 |
+
"loss": 0.019,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.5069124423963134,
|
| 315 |
+
"grad_norm": 0.3057345449924469,
|
| 316 |
+
"learning_rate": 6.402892702827916e-05,
|
| 317 |
+
"loss": 0.0204,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5184331797235023,
|
| 322 |
+
"grad_norm": 0.27462145686149597,
|
| 323 |
+
"learning_rate": 6.243449435824276e-05,
|
| 324 |
+
"loss": 0.0198,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5299539170506913,
|
| 329 |
+
"grad_norm": 0.2001851201057434,
|
| 330 |
+
"learning_rate": 6.0826464751186994e-05,
|
| 331 |
+
"loss": 0.0189,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5414746543778802,
|
| 336 |
+
"grad_norm": 0.2672295570373535,
|
| 337 |
+
"learning_rate": 5.9206596563878357e-05,
|
| 338 |
+
"loss": 0.0178,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5529953917050692,
|
| 343 |
+
"grad_norm": 0.3161514699459076,
|
| 344 |
+
"learning_rate": 5.757666109839702e-05,
|
| 345 |
+
"loss": 0.0213,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5645161290322581,
|
| 350 |
+
"grad_norm": 0.22500912845134735,
|
| 351 |
+
"learning_rate": 5.5938440665244006e-05,
|
| 352 |
+
"loss": 0.0191,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.576036866359447,
|
| 357 |
+
"grad_norm": 0.32964590191841125,
|
| 358 |
+
"learning_rate": 5.4293726634410855e-05,
|
| 359 |
+
"loss": 0.0194,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5875576036866359,
|
| 364 |
+
"grad_norm": 0.2184433490037918,
|
| 365 |
+
"learning_rate": 5.264431747654284e-05,
|
| 366 |
+
"loss": 0.0181,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5990783410138248,
|
| 371 |
+
"grad_norm": 0.22752192616462708,
|
| 372 |
+
"learning_rate": 5.0992016796337686e-05,
|
| 373 |
+
"loss": 0.0153,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6105990783410138,
|
| 378 |
+
"grad_norm": 0.17904232442378998,
|
| 379 |
+
"learning_rate": 4.93386313603304e-05,
|
| 380 |
+
"loss": 0.017,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6221198156682027,
|
| 385 |
+
"grad_norm": 0.30245211720466614,
|
| 386 |
+
"learning_rate": 4.7685969121220456e-05,
|
| 387 |
+
"loss": 0.017,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6336405529953917,
|
| 392 |
+
"grad_norm": 0.23163466155529022,
|
| 393 |
+
"learning_rate": 4.60358372409022e-05,
|
| 394 |
+
"loss": 0.0181,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6451612903225806,
|
| 399 |
+
"grad_norm": 0.22935254871845245,
|
| 400 |
+
"learning_rate": 4.439004011435979e-05,
|
| 401 |
+
"loss": 0.0181,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6566820276497696,
|
| 406 |
+
"grad_norm": 0.220436692237854,
|
| 407 |
+
"learning_rate": 4.275037739658771e-05,
|
| 408 |
+
"loss": 0.017,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6682027649769585,
|
| 413 |
+
"grad_norm": 0.26229238510131836,
|
| 414 |
+
"learning_rate": 4.111864203469457e-05,
|
| 415 |
+
"loss": 0.0178,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6797235023041475,
|
| 420 |
+
"grad_norm": 0.19217033684253693,
|
| 421 |
+
"learning_rate": 3.949661830734172e-05,
|
| 422 |
+
"loss": 0.0152,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6912442396313364,
|
| 427 |
+
"grad_norm": 0.20372073352336884,
|
| 428 |
+
"learning_rate": 3.788607987366069e-05,
|
| 429 |
+
"loss": 0.0159,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.7027649769585254,
|
| 434 |
+
"grad_norm": 0.1933118999004364,
|
| 435 |
+
"learning_rate": 3.628878783378302e-05,
|
| 436 |
+
"loss": 0.0157,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.7142857142857143,
|
| 441 |
+
"grad_norm": 0.19851423799991608,
|
| 442 |
+
"learning_rate": 3.470648880310313e-05,
|
| 443 |
+
"loss": 0.0145,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.7258064516129032,
|
| 448 |
+
"grad_norm": 0.16878198087215424,
|
| 449 |
+
"learning_rate": 3.3140913002379995e-05,
|
| 450 |
+
"loss": 0.0157,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.7373271889400922,
|
| 455 |
+
"grad_norm": 0.21986301243305206,
|
| 456 |
+
"learning_rate": 3.1593772365766105e-05,
|
| 457 |
+
"loss": 0.0162,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.7488479262672811,
|
| 462 |
+
"grad_norm": 0.13134035468101501,
|
| 463 |
+
"learning_rate": 3.006675866883275e-05,
|
| 464 |
+
"loss": 0.0157,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.7603686635944701,
|
| 469 |
+
"grad_norm": 0.1785222440958023,
|
| 470 |
+
"learning_rate": 2.8561541678638142e-05,
|
| 471 |
+
"loss": 0.0147,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.771889400921659,
|
| 476 |
+
"grad_norm": 0.16605904698371887,
|
| 477 |
+
"learning_rate": 2.707976732786166e-05,
|
| 478 |
+
"loss": 0.0133,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.783410138248848,
|
| 483 |
+
"grad_norm": 0.18675336241722107,
|
| 484 |
+
"learning_rate": 2.562305591500069e-05,
|
| 485 |
+
"loss": 0.0145,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.7949308755760369,
|
| 490 |
+
"grad_norm": 0.18020185828208923,
|
| 491 |
+
"learning_rate": 2.419300033259798e-05,
|
| 492 |
+
"loss": 0.0151,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8064516129032258,
|
| 497 |
+
"grad_norm": 0.1856432557106018,
|
| 498 |
+
"learning_rate": 2.279116432543705e-05,
|
| 499 |
+
"loss": 0.0137,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.8179723502304147,
|
| 504 |
+
"grad_norm": 0.1251407414674759,
|
| 505 |
+
"learning_rate": 2.1419080780610123e-05,
|
| 506 |
+
"loss": 0.0126,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.8294930875576036,
|
| 511 |
+
"grad_norm": 0.20187409222126007,
|
| 512 |
+
"learning_rate": 2.0078250051328784e-05,
|
| 513 |
+
"loss": 0.0139,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.8410138248847926,
|
| 518 |
+
"grad_norm": 0.16873343288898468,
|
| 519 |
+
"learning_rate": 1.877013831630961e-05,
|
| 520 |
+
"loss": 0.0142,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.8525345622119815,
|
| 525 |
+
"grad_norm": 0.11382901668548584,
|
| 526 |
+
"learning_rate": 1.749617597652934e-05,
|
| 527 |
+
"loss": 0.0129,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.8640552995391705,
|
| 532 |
+
"grad_norm": 0.1710187941789627,
|
| 533 |
+
"learning_rate": 1.62577560911024e-05,
|
| 534 |
+
"loss": 0.0152,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8755760368663594,
|
| 539 |
+
"grad_norm": 0.12709008157253265,
|
| 540 |
+
"learning_rate": 1.5056232853991209e-05,
|
| 541 |
+
"loss": 0.0139,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8870967741935484,
|
| 546 |
+
"grad_norm": 0.15011294186115265,
|
| 547 |
+
"learning_rate": 1.389292011321498e-05,
|
| 548 |
+
"loss": 0.0136,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8986175115207373,
|
| 553 |
+
"grad_norm": 0.17724603414535522,
|
| 554 |
+
"learning_rate": 1.2769089934176126e-05,
|
| 555 |
+
"loss": 0.0125,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9101382488479263,
|
| 560 |
+
"grad_norm": 0.16647891700267792,
|
| 561 |
+
"learning_rate": 1.1685971208675539e-05,
|
| 562 |
+
"loss": 0.0129,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.9216589861751152,
|
| 567 |
+
"grad_norm": 0.1402997523546219,
|
| 568 |
+
"learning_rate": 1.0644748311137376e-05,
|
| 569 |
+
"loss": 0.0123,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.9331797235023042,
|
| 574 |
+
"grad_norm": 0.11741903424263,
|
| 575 |
+
"learning_rate": 9.646559803512994e-06,
|
| 576 |
+
"loss": 0.0119,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.9447004608294931,
|
| 581 |
+
"grad_norm": 0.14751701056957245,
|
| 582 |
+
"learning_rate": 8.692497190280224e-06,
|
| 583 |
+
"loss": 0.0133,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.956221198156682,
|
| 588 |
+
"grad_norm": 0.14610819518566132,
|
| 589 |
+
"learning_rate": 7.783603724899257e-06,
|
| 590 |
+
"loss": 0.0128,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.967741935483871,
|
| 595 |
+
"grad_norm": 0.1284884661436081,
|
| 596 |
+
"learning_rate": 6.92087326903022e-06,
|
| 597 |
+
"loss": 0.0116,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.9792626728110599,
|
| 602 |
+
"grad_norm": 0.14618724584579468,
|
| 603 |
+
"learning_rate": 6.1052492057601275e-06,
|
| 604 |
+
"loss": 0.0137,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9907834101382489,
|
| 609 |
+
"grad_norm": 0.14276568591594696,
|
| 610 |
+
"learning_rate": 5.337623408027293e-06,
|
| 611 |
+
"loss": 0.0136,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.0023041474654377,
|
| 616 |
+
"grad_norm": 0.10031285136938095,
|
| 617 |
+
"learning_rate": 4.618835263371396e-06,
|
| 618 |
+
"loss": 0.0119,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.0138248847926268,
|
| 623 |
+
"grad_norm": 0.10763294994831085,
|
| 624 |
+
"learning_rate": 3.949670756075447e-06,
|
| 625 |
+
"loss": 0.0111,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.0253456221198156,
|
| 630 |
+
"grad_norm": 0.1076810285449028,
|
| 631 |
+
"learning_rate": 3.3308616077036115e-06,
|
| 632 |
+
"loss": 0.0109,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.0368663594470047,
|
| 637 |
+
"grad_norm": 0.13353325426578522,
|
| 638 |
+
"learning_rate": 2.7630844769743757e-06,
|
| 639 |
+
"loss": 0.011,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.0483870967741935,
|
| 644 |
+
"grad_norm": 0.10967040807008743,
|
| 645 |
+
"learning_rate": 2.2469602198441573e-06,
|
| 646 |
+
"loss": 0.0117,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.0599078341013826,
|
| 651 |
+
"grad_norm": 0.16759690642356873,
|
| 652 |
+
"learning_rate": 1.7830532106104747e-06,
|
| 653 |
+
"loss": 0.0118,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.0714285714285714,
|
| 658 |
+
"grad_norm": 0.15041400492191315,
|
| 659 |
+
"learning_rate": 1.3718707247769135e-06,
|
| 660 |
+
"loss": 0.0121,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.0829493087557605,
|
| 665 |
+
"grad_norm": 0.11416321247816086,
|
| 666 |
+
"learning_rate": 1.0138623843548078e-06,
|
| 667 |
+
"loss": 0.0127,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.0944700460829493,
|
| 672 |
+
"grad_norm": 0.12047427892684937,
|
| 673 |
+
"learning_rate": 7.094196662081831e-07,
|
| 674 |
+
"loss": 0.0117,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.1059907834101383,
|
| 679 |
+
"grad_norm": 0.13395977020263672,
|
| 680 |
+
"learning_rate": 4.5887547397955864e-07,
|
| 681 |
+
"loss": 0.0108,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.1175115207373272,
|
| 686 |
+
"grad_norm": 0.08437898010015488,
|
| 687 |
+
"learning_rate": 2.625037740646763e-07,
|
| 688 |
+
"loss": 0.0123,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.129032258064516,
|
| 693 |
+
"grad_norm": 0.07629328966140747,
|
| 694 |
+
"learning_rate": 1.2051929603428825e-07,
|
| 695 |
+
"loss": 0.0111,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.140552995391705,
|
| 700 |
+
"grad_norm": 0.10458842664957047,
|
| 701 |
+
"learning_rate": 3.3077297830541584e-08,
|
| 702 |
+
"loss": 0.0128,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.1520737327188941,
|
| 707 |
+
"grad_norm": 0.11208489537239075,
|
| 708 |
+
"learning_rate": 2.7339599464326627e-10,
|
| 709 |
+
"loss": 0.0123,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.1520737327188941,
|
| 714 |
+
"step": 1000,
|
| 715 |
+
"total_flos": 0.0,
|
| 716 |
+
"train_loss": 0.03519474593549967,
|
| 717 |
+
"train_runtime": 1870.4299,
|
| 718 |
+
"train_samples_per_second": 53.464,
|
| 719 |
+
"train_steps_per_second": 0.535
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 10,
|
| 723 |
+
"max_steps": 1000,
|
| 724 |
+
"num_input_tokens_seen": 0,
|
| 725 |
+
"num_train_epochs": 2,
|
| 726 |
+
"save_steps": 1000,
|
| 727 |
+
"stateful_callbacks": {
|
| 728 |
+
"TrainerControl": {
|
| 729 |
+
"args": {
|
| 730 |
+
"should_epoch_stop": false,
|
| 731 |
+
"should_evaluate": false,
|
| 732 |
+
"should_log": false,
|
| 733 |
+
"should_save": true,
|
| 734 |
+
"should_training_stop": true
|
| 735 |
+
},
|
| 736 |
+
"attributes": {}
|
| 737 |
+
}
|
| 738 |
+
},
|
| 739 |
+
"total_flos": 0.0,
|
| 740 |
+
"train_batch_size": 100,
|
| 741 |
+
"trial_name": null,
|
| 742 |
+
"trial_params": null
|
| 743 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e0ae2c3a8057ef71b3d22f0c8dd728c42aa8b67ed4242ad7cc20cd753eb2a04
|
| 3 |
+
size 5304
|