Upload folder using huggingface_hub
Browse files- config.json +64 -0
- experiment_cfg/metadata.json +355 -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 +1,355 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"new_embodiment": {
|
| 3 |
+
"statistics": {
|
| 4 |
+
"state": {
|
| 5 |
+
"left_arm": {
|
| 6 |
+
"max": [
|
| 7 |
+
34.16289520263672,
|
| 8 |
+
61.91222381591797,
|
| 9 |
+
99.38703918457031,
|
| 10 |
+
75.7874755859375,
|
| 11 |
+
46.27594757080078
|
| 12 |
+
],
|
| 13 |
+
"min": [
|
| 14 |
+
-27.90346908569336,
|
| 15 |
+
-99.60814666748047,
|
| 16 |
+
-33.27495574951172,
|
| 17 |
+
-88.31119537353516,
|
| 18 |
+
-66.83760833740234
|
| 19 |
+
],
|
| 20 |
+
"mean": [
|
| 21 |
+
-1.3637042045593262,
|
| 22 |
+
-70.07746124267578,
|
| 23 |
+
81.53594970703125,
|
| 24 |
+
46.74204635620117,
|
| 25 |
+
-32.77935791015625
|
| 26 |
+
],
|
| 27 |
+
"std": [
|
| 28 |
+
10.373152732849121,
|
| 29 |
+
52.60585021972656,
|
| 30 |
+
34.18480682373047,
|
| 31 |
+
48.48912811279297,
|
| 32 |
+
37.322959899902344
|
| 33 |
+
],
|
| 34 |
+
"q01": [
|
| 35 |
+
-22.021116256713867,
|
| 36 |
+
-99.52977752685547,
|
| 37 |
+
-18.82662010192871,
|
| 38 |
+
-66.29981231689453,
|
| 39 |
+
-58.38827896118164
|
| 40 |
+
],
|
| 41 |
+
"q99": [
|
| 42 |
+
26.69683265686035,
|
| 43 |
+
51.0971794128418,
|
| 44 |
+
98.94921112060547,
|
| 45 |
+
74.72486114501953,
|
| 46 |
+
45.88522720336914
|
| 47 |
+
]
|
| 48 |
+
},
|
| 49 |
+
"gripper1": {
|
| 50 |
+
"max": [
|
| 51 |
+
71.77835083007812
|
| 52 |
+
],
|
| 53 |
+
"min": [
|
| 54 |
+
0.0
|
| 55 |
+
],
|
| 56 |
+
"mean": [
|
| 57 |
+
10.566564559936523
|
| 58 |
+
],
|
| 59 |
+
"std": [
|
| 60 |
+
10.46413803100586
|
| 61 |
+
],
|
| 62 |
+
"q01": [
|
| 63 |
+
0.4510309398174286
|
| 64 |
+
],
|
| 65 |
+
"q99": [
|
| 66 |
+
56.572166442871094
|
| 67 |
+
]
|
| 68 |
+
},
|
| 69 |
+
"right_arm": {
|
| 70 |
+
"max": [
|
| 71 |
+
31.229597091674805,
|
| 72 |
+
73.70741271972656,
|
| 73 |
+
98.10017395019531,
|
| 74 |
+
84.24878692626953,
|
| 75 |
+
41.3212776184082
|
| 76 |
+
],
|
| 77 |
+
"min": [
|
| 78 |
+
-27.022125244140625,
|
| 79 |
+
-97.75550842285156,
|
| 80 |
+
-63.21243667602539,
|
| 81 |
+
-96.60742950439453,
|
| 82 |
+
-65.18462371826172
|
| 83 |
+
],
|
| 84 |
+
"mean": [
|
| 85 |
+
-1.5135765075683594,
|
| 86 |
+
-25.33967399597168,
|
| 87 |
+
61.376529693603516,
|
| 88 |
+
0.33546876907348633,
|
| 89 |
+
-41.29887390136719
|
| 90 |
+
],
|
| 91 |
+
"std": [
|
| 92 |
+
12.458887100219727,
|
| 93 |
+
60.93878936767578,
|
| 94 |
+
35.7954216003418,
|
| 95 |
+
60.31285858154297,
|
| 96 |
+
23.256187438964844
|
| 97 |
+
],
|
| 98 |
+
"q01": [
|
| 99 |
+
-22.887195587158203,
|
| 100 |
+
-97.51502990722656,
|
| 101 |
+
-18.134714126586914,
|
| 102 |
+
-95.5573501586914,
|
| 103 |
+
-57.9000244140625
|
| 104 |
+
],
|
| 105 |
+
"q99": [
|
| 106 |
+
28.61806297302246,
|
| 107 |
+
63.446895599365234,
|
| 108 |
+
97.84110260009766,
|
| 109 |
+
73.2633285522461,
|
| 110 |
+
29.41472053527832
|
| 111 |
+
]
|
| 112 |
+
},
|
| 113 |
+
"gripper2": {
|
| 114 |
+
"max": [
|
| 115 |
+
75.05668640136719
|
| 116 |
+
],
|
| 117 |
+
"min": [
|
| 118 |
+
2.324263095855713
|
| 119 |
+
],
|
| 120 |
+
"mean": [
|
| 121 |
+
23.147193908691406
|
| 122 |
+
],
|
| 123 |
+
"std": [
|
| 124 |
+
13.153349876403809
|
| 125 |
+
],
|
| 126 |
+
"q01": [
|
| 127 |
+
2.551020383834839
|
| 128 |
+
],
|
| 129 |
+
"q99": [
|
| 130 |
+
61.621315002441406
|
| 131 |
+
]
|
| 132 |
+
}
|
| 133 |
+
},
|
| 134 |
+
"action": {
|
| 135 |
+
"left_arm": {
|
| 136 |
+
"max": [
|
| 137 |
+
35.92695617675781,
|
| 138 |
+
62.01413345336914,
|
| 139 |
+
100.0,
|
| 140 |
+
76.97608184814453,
|
| 141 |
+
46.617828369140625
|
| 142 |
+
],
|
| 143 |
+
"min": [
|
| 144 |
+
-28.066692352294922,
|
| 145 |
+
-100.0,
|
| 146 |
+
-35.07326126098633,
|
| 147 |
+
-89.78516387939453,
|
| 148 |
+
-67.17948913574219
|
| 149 |
+
],
|
| 150 |
+
"mean": [
|
| 151 |
+
-1.1567177772521973,
|
| 152 |
+
-70.6091537475586,
|
| 153 |
+
82.1029281616211,
|
| 154 |
+
47.767425537109375,
|
| 155 |
+
-32.80006408691406
|
| 156 |
+
],
|
| 157 |
+
"std": [
|
| 158 |
+
10.52210521697998,
|
| 159 |
+
52.541717529296875,
|
| 160 |
+
35.224395751953125,
|
| 161 |
+
49.508544921875,
|
| 162 |
+
37.26472473144531
|
| 163 |
+
],
|
| 164 |
+
"q01": [
|
| 165 |
+
-21.873760223388672,
|
| 166 |
+
-100.0,
|
| 167 |
+
-20.238094329833984,
|
| 168 |
+
-66.76124572753906,
|
| 169 |
+
-58.43711853027344
|
| 170 |
+
],
|
| 171 |
+
"q99": [
|
| 172 |
+
27.669710159301758,
|
| 173 |
+
50.230565567016654,
|
| 174 |
+
100.0,
|
| 175 |
+
76.97608184814453,
|
| 176 |
+
45.738704681396484
|
| 177 |
+
]
|
| 178 |
+
},
|
| 179 |
+
"gripper1": {
|
| 180 |
+
"max": [
|
| 181 |
+
74.2166519165039
|
| 182 |
+
],
|
| 183 |
+
"min": [
|
| 184 |
+
0.0
|
| 185 |
+
],
|
| 186 |
+
"mean": [
|
| 187 |
+
10.451210021972656
|
| 188 |
+
],
|
| 189 |
+
"std": [
|
| 190 |
+
10.781113624572754
|
| 191 |
+
],
|
| 192 |
+
"q01": [
|
| 193 |
+
0.08952551335096359
|
| 194 |
+
],
|
| 195 |
+
"q99": [
|
| 196 |
+
56.938228607177734
|
| 197 |
+
]
|
| 198 |
+
},
|
| 199 |
+
"right_arm": {
|
| 200 |
+
"max": [
|
| 201 |
+
31.447315216064453,
|
| 202 |
+
72.47631072998047,
|
| 203 |
+
99.20035552978516,
|
| 204 |
+
85.42713928222656,
|
| 205 |
+
42.69340896606445
|
| 206 |
+
],
|
| 207 |
+
"min": [
|
| 208 |
+
-27.511274337768555,
|
| 209 |
+
-99.25834655761719,
|
| 210 |
+
-67.30342102050781,
|
| 211 |
+
-98.40870666503906,
|
| 212 |
+
-66.81427764892578
|
| 213 |
+
],
|
| 214 |
+
"mean": [
|
| 215 |
+
-1.471267580986023,
|
| 216 |
+
-26.692821502685547,
|
| 217 |
+
60.74913787841797,
|
| 218 |
+
0.23357370495796204,
|
| 219 |
+
-41.19910430908203
|
| 220 |
+
],
|
| 221 |
+
"std": [
|
| 222 |
+
12.504396438598633,
|
| 223 |
+
60.0944709777832,
|
| 224 |
+
36.86882400512695,
|
| 225 |
+
60.7491340637207,
|
| 226 |
+
23.23846435546875
|
| 227 |
+
],
|
| 228 |
+
"q01": [
|
| 229 |
+
-23.001230239868164,
|
| 230 |
+
-97.692626953125,
|
| 231 |
+
-19.68014144897461,
|
| 232 |
+
-96.31491088867188,
|
| 233 |
+
-57.749412536621094
|
| 234 |
+
],
|
| 235 |
+
"q99": [
|
| 236 |
+
28.823287963867188,
|
| 237 |
+
62.34033966064453,
|
| 238 |
+
98.57840728759766,
|
| 239 |
+
74.03684997558594,
|
| 240 |
+
29.721281051635742
|
| 241 |
+
]
|
| 242 |
+
},
|
| 243 |
+
"gripper2": {
|
| 244 |
+
"max": [
|
| 245 |
+
77.73826599121094
|
| 246 |
+
],
|
| 247 |
+
"min": [
|
| 248 |
+
0.4267425239086151
|
| 249 |
+
],
|
| 250 |
+
"mean": [
|
| 251 |
+
21.515247344970703
|
| 252 |
+
],
|
| 253 |
+
"std": [
|
| 254 |
+
14.212292671203613
|
| 255 |
+
],
|
| 256 |
+
"q01": [
|
| 257 |
+
1.8492176532745361
|
| 258 |
+
],
|
| 259 |
+
"q99": [
|
| 260 |
+
61.87766647338867
|
| 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 |
+
"top_rgb": {
|
| 276 |
+
"resolution": [
|
| 277 |
+
640,
|
| 278 |
+
480
|
| 279 |
+
],
|
| 280 |
+
"channels": 3,
|
| 281 |
+
"fps": 30.0
|
| 282 |
+
}
|
| 283 |
+
},
|
| 284 |
+
"state": {
|
| 285 |
+
"left_arm": {
|
| 286 |
+
"absolute": true,
|
| 287 |
+
"rotation_type": null,
|
| 288 |
+
"shape": [
|
| 289 |
+
5
|
| 290 |
+
],
|
| 291 |
+
"continuous": true
|
| 292 |
+
},
|
| 293 |
+
"gripper1": {
|
| 294 |
+
"absolute": true,
|
| 295 |
+
"rotation_type": null,
|
| 296 |
+
"shape": [
|
| 297 |
+
1
|
| 298 |
+
],
|
| 299 |
+
"continuous": true
|
| 300 |
+
},
|
| 301 |
+
"right_arm": {
|
| 302 |
+
"absolute": true,
|
| 303 |
+
"rotation_type": null,
|
| 304 |
+
"shape": [
|
| 305 |
+
5
|
| 306 |
+
],
|
| 307 |
+
"continuous": true
|
| 308 |
+
},
|
| 309 |
+
"gripper2": {
|
| 310 |
+
"absolute": true,
|
| 311 |
+
"rotation_type": null,
|
| 312 |
+
"shape": [
|
| 313 |
+
1
|
| 314 |
+
],
|
| 315 |
+
"continuous": true
|
| 316 |
+
}
|
| 317 |
+
},
|
| 318 |
+
"action": {
|
| 319 |
+
"left_arm": {
|
| 320 |
+
"absolute": true,
|
| 321 |
+
"rotation_type": null,
|
| 322 |
+
"shape": [
|
| 323 |
+
5
|
| 324 |
+
],
|
| 325 |
+
"continuous": true
|
| 326 |
+
},
|
| 327 |
+
"gripper1": {
|
| 328 |
+
"absolute": true,
|
| 329 |
+
"rotation_type": null,
|
| 330 |
+
"shape": [
|
| 331 |
+
1
|
| 332 |
+
],
|
| 333 |
+
"continuous": true
|
| 334 |
+
},
|
| 335 |
+
"right_arm": {
|
| 336 |
+
"absolute": true,
|
| 337 |
+
"rotation_type": null,
|
| 338 |
+
"shape": [
|
| 339 |
+
5
|
| 340 |
+
],
|
| 341 |
+
"continuous": true
|
| 342 |
+
},
|
| 343 |
+
"gripper2": {
|
| 344 |
+
"absolute": true,
|
| 345 |
+
"rotation_type": null,
|
| 346 |
+
"shape": [
|
| 347 |
+
1
|
| 348 |
+
],
|
| 349 |
+
"continuous": true
|
| 350 |
+
}
|
| 351 |
+
}
|
| 352 |
+
},
|
| 353 |
+
"embodiment_tag": "new_embodiment"
|
| 354 |
+
}
|
| 355 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:115fd57ab6985556b75222944abe6f1f152a92a06ef614f5756448fabd7c02a8
|
| 3 |
+
size 4999367032
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38e59620503cd8a0523f4851324145e9f86aba1a919018b013f473ad5324b370
|
| 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": 0.9025270758122743,
|
| 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": 0.009025270758122744,
|
| 14 |
+
"grad_norm": 1.977829933166504,
|
| 15 |
+
"learning_rate": 1.8e-05,
|
| 16 |
+
"loss": 0.9407,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.018050541516245487,
|
| 21 |
+
"grad_norm": 0.9880571365356445,
|
| 22 |
+
"learning_rate": 3.8e-05,
|
| 23 |
+
"loss": 0.2558,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02707581227436823,
|
| 28 |
+
"grad_norm": 1.016868233680725,
|
| 29 |
+
"learning_rate": 5.8e-05,
|
| 30 |
+
"loss": 0.151,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.036101083032490974,
|
| 35 |
+
"grad_norm": 1.0683897733688354,
|
| 36 |
+
"learning_rate": 7.800000000000001e-05,
|
| 37 |
+
"loss": 0.1332,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04512635379061372,
|
| 42 |
+
"grad_norm": 0.8018341660499573,
|
| 43 |
+
"learning_rate": 9.8e-05,
|
| 44 |
+
"loss": 0.0983,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05415162454873646,
|
| 49 |
+
"grad_norm": 0.6524431109428406,
|
| 50 |
+
"learning_rate": 9.997785653888835e-05,
|
| 51 |
+
"loss": 0.0808,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0631768953068592,
|
| 56 |
+
"grad_norm": 0.6845595240592957,
|
| 57 |
+
"learning_rate": 9.990133642141359e-05,
|
| 58 |
+
"loss": 0.0778,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07220216606498195,
|
| 63 |
+
"grad_norm": 0.6983832120895386,
|
| 64 |
+
"learning_rate": 9.977024992520602e-05,
|
| 65 |
+
"loss": 0.0711,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0812274368231047,
|
| 70 |
+
"grad_norm": 0.6112372279167175,
|
| 71 |
+
"learning_rate": 9.95847403914247e-05,
|
| 72 |
+
"loss": 0.0679,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.09025270758122744,
|
| 77 |
+
"grad_norm": 0.5820932388305664,
|
| 78 |
+
"learning_rate": 9.934501067202117e-05,
|
| 79 |
+
"loss": 0.0641,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09927797833935018,
|
| 84 |
+
"grad_norm": 0.5818979144096375,
|
| 85 |
+
"learning_rate": 9.905132290792394e-05,
|
| 86 |
+
"loss": 0.0618,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10830324909747292,
|
| 91 |
+
"grad_norm": 0.6202858686447144,
|
| 92 |
+
"learning_rate": 9.870399824239117e-05,
|
| 93 |
+
"loss": 0.0554,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.11732851985559567,
|
| 98 |
+
"grad_norm": 0.5585798621177673,
|
| 99 |
+
"learning_rate": 9.830341646984521e-05,
|
| 100 |
+
"loss": 0.0559,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.1263537906137184,
|
| 105 |
+
"grad_norm": 0.6366000771522522,
|
| 106 |
+
"learning_rate": 9.785001562057309e-05,
|
| 107 |
+
"loss": 0.0553,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.13537906137184116,
|
| 112 |
+
"grad_norm": 0.4224783778190613,
|
| 113 |
+
"learning_rate": 9.734429148174675e-05,
|
| 114 |
+
"loss": 0.05,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1444043321299639,
|
| 119 |
+
"grad_norm": 0.39184054732322693,
|
| 120 |
+
"learning_rate": 9.6786797055287e-05,
|
| 121 |
+
"loss": 0.0455,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.15342960288808663,
|
| 126 |
+
"grad_norm": 0.4573412835597992,
|
| 127 |
+
"learning_rate": 9.617814195316411e-05,
|
| 128 |
+
"loss": 0.0408,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1624548736462094,
|
| 133 |
+
"grad_norm": 0.4405965805053711,
|
| 134 |
+
"learning_rate": 9.551899173079607e-05,
|
| 135 |
+
"loss": 0.0455,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.17148014440433212,
|
| 140 |
+
"grad_norm": 0.46750208735466003,
|
| 141 |
+
"learning_rate": 9.481006715927351e-05,
|
| 142 |
+
"loss": 0.0448,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.18050541516245489,
|
| 147 |
+
"grad_norm": 0.37655723094940186,
|
| 148 |
+
"learning_rate": 9.405214343720707e-05,
|
| 149 |
+
"loss": 0.0439,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18953068592057762,
|
| 154 |
+
"grad_norm": 0.5338054895401001,
|
| 155 |
+
"learning_rate": 9.32460493430591e-05,
|
| 156 |
+
"loss": 0.047,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.19855595667870035,
|
| 161 |
+
"grad_norm": 0.4981560707092285,
|
| 162 |
+
"learning_rate": 9.239266632888659e-05,
|
| 163 |
+
"loss": 0.0418,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.2075812274368231,
|
| 168 |
+
"grad_norm": 0.4395357668399811,
|
| 169 |
+
"learning_rate": 9.14929275564863e-05,
|
| 170 |
+
"loss": 0.0408,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.21660649819494585,
|
| 175 |
+
"grad_norm": 0.44028720259666443,
|
| 176 |
+
"learning_rate": 9.0547816876996e-05,
|
| 177 |
+
"loss": 0.0409,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.22563176895306858,
|
| 182 |
+
"grad_norm": 0.37459373474121094,
|
| 183 |
+
"learning_rate": 8.955836775506776e-05,
|
| 184 |
+
"loss": 0.0426,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.23465703971119134,
|
| 189 |
+
"grad_norm": 0.4326346516609192,
|
| 190 |
+
"learning_rate": 8.852566213878947e-05,
|
| 191 |
+
"loss": 0.0429,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.24368231046931407,
|
| 196 |
+
"grad_norm": 0.45173099637031555,
|
| 197 |
+
"learning_rate": 8.745082927659047e-05,
|
| 198 |
+
"loss": 0.0384,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.2527075812274368,
|
| 203 |
+
"grad_norm": 0.33228954672813416,
|
| 204 |
+
"learning_rate": 8.633504448242505e-05,
|
| 205 |
+
"loss": 0.0422,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.26173285198555957,
|
| 210 |
+
"grad_norm": 0.364307165145874,
|
| 211 |
+
"learning_rate": 8.517952785058385e-05,
|
| 212 |
+
"loss": 0.0369,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.27075812274368233,
|
| 217 |
+
"grad_norm": 0.34387287497520447,
|
| 218 |
+
"learning_rate": 8.398554292153866e-05,
|
| 219 |
+
"loss": 0.0353,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.27978339350180503,
|
| 224 |
+
"grad_norm": 0.3851270079612732,
|
| 225 |
+
"learning_rate": 8.275439530027948e-05,
|
| 226 |
+
"loss": 0.0417,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2888086642599278,
|
| 231 |
+
"grad_norm": 0.28114214539527893,
|
| 232 |
+
"learning_rate": 8.148743122865463e-05,
|
| 233 |
+
"loss": 0.0345,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.29783393501805056,
|
| 238 |
+
"grad_norm": 0.33744707703590393,
|
| 239 |
+
"learning_rate": 8.018603611327504e-05,
|
| 240 |
+
"loss": 0.0342,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.30685920577617326,
|
| 245 |
+
"grad_norm": 0.4103146195411682,
|
| 246 |
+
"learning_rate": 7.88516330105925e-05,
|
| 247 |
+
"loss": 0.035,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.315884476534296,
|
| 252 |
+
"grad_norm": 0.328669011592865,
|
| 253 |
+
"learning_rate": 7.748568107080832e-05,
|
| 254 |
+
"loss": 0.0327,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3249097472924188,
|
| 259 |
+
"grad_norm": 0.4371018409729004,
|
| 260 |
+
"learning_rate": 7.608967394231387e-05,
|
| 261 |
+
"loss": 0.0359,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.33393501805054154,
|
| 266 |
+
"grad_norm": 0.4694584012031555,
|
| 267 |
+
"learning_rate": 7.466513813840825e-05,
|
| 268 |
+
"loss": 0.0339,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.34296028880866425,
|
| 273 |
+
"grad_norm": 0.38598158955574036,
|
| 274 |
+
"learning_rate": 7.32136313680782e-05,
|
| 275 |
+
"loss": 0.0289,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.351985559566787,
|
| 280 |
+
"grad_norm": 0.36169150471687317,
|
| 281 |
+
"learning_rate": 7.173674083266624e-05,
|
| 282 |
+
"loss": 0.032,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.36101083032490977,
|
| 287 |
+
"grad_norm": 0.3175145983695984,
|
| 288 |
+
"learning_rate": 7.023608149028937e-05,
|
| 289 |
+
"loss": 0.0311,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.3700361010830325,
|
| 294 |
+
"grad_norm": 0.4808519184589386,
|
| 295 |
+
"learning_rate": 6.871329428990602e-05,
|
| 296 |
+
"loss": 0.0351,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.37906137184115524,
|
| 301 |
+
"grad_norm": 0.3936750590801239,
|
| 302 |
+
"learning_rate": 6.71700443769625e-05,
|
| 303 |
+
"loss": 0.0297,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.388086642599278,
|
| 308 |
+
"grad_norm": 0.31872764229774475,
|
| 309 |
+
"learning_rate": 6.56080192725808e-05,
|
| 310 |
+
"loss": 0.029,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3971119133574007,
|
| 315 |
+
"grad_norm": 0.3199341297149658,
|
| 316 |
+
"learning_rate": 6.402892702827916e-05,
|
| 317 |
+
"loss": 0.0305,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.40613718411552346,
|
| 322 |
+
"grad_norm": 0.29824522137641907,
|
| 323 |
+
"learning_rate": 6.243449435824276e-05,
|
| 324 |
+
"loss": 0.0313,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4151624548736462,
|
| 329 |
+
"grad_norm": 0.3850104808807373,
|
| 330 |
+
"learning_rate": 6.0826464751186994e-05,
|
| 331 |
+
"loss": 0.0307,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.42418772563176893,
|
| 336 |
+
"grad_norm": 0.28914761543273926,
|
| 337 |
+
"learning_rate": 5.9206596563878357e-05,
|
| 338 |
+
"loss": 0.0335,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.4332129963898917,
|
| 343 |
+
"grad_norm": 0.29466763138771057,
|
| 344 |
+
"learning_rate": 5.757666109839702e-05,
|
| 345 |
+
"loss": 0.0282,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.44223826714801445,
|
| 350 |
+
"grad_norm": 0.3107747435569763,
|
| 351 |
+
"learning_rate": 5.5938440665244006e-05,
|
| 352 |
+
"loss": 0.027,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.45126353790613716,
|
| 357 |
+
"grad_norm": 0.38809922337532043,
|
| 358 |
+
"learning_rate": 5.4293726634410855e-05,
|
| 359 |
+
"loss": 0.0281,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4602888086642599,
|
| 364 |
+
"grad_norm": 0.27953779697418213,
|
| 365 |
+
"learning_rate": 5.264431747654284e-05,
|
| 366 |
+
"loss": 0.0251,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4693140794223827,
|
| 371 |
+
"grad_norm": 0.24429909884929657,
|
| 372 |
+
"learning_rate": 5.0992016796337686e-05,
|
| 373 |
+
"loss": 0.0258,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.47833935018050544,
|
| 378 |
+
"grad_norm": 0.21034276485443115,
|
| 379 |
+
"learning_rate": 4.93386313603304e-05,
|
| 380 |
+
"loss": 0.0271,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.48736462093862815,
|
| 385 |
+
"grad_norm": 0.29745611548423767,
|
| 386 |
+
"learning_rate": 4.7685969121220456e-05,
|
| 387 |
+
"loss": 0.0268,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4963898916967509,
|
| 392 |
+
"grad_norm": 0.2871965765953064,
|
| 393 |
+
"learning_rate": 4.60358372409022e-05,
|
| 394 |
+
"loss": 0.0245,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5054151624548736,
|
| 399 |
+
"grad_norm": 0.30364564061164856,
|
| 400 |
+
"learning_rate": 4.439004011435979e-05,
|
| 401 |
+
"loss": 0.0237,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5144404332129964,
|
| 406 |
+
"grad_norm": 0.28791144490242004,
|
| 407 |
+
"learning_rate": 4.275037739658771e-05,
|
| 408 |
+
"loss": 0.0264,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5234657039711191,
|
| 413 |
+
"grad_norm": 0.3472572863101959,
|
| 414 |
+
"learning_rate": 4.111864203469457e-05,
|
| 415 |
+
"loss": 0.0268,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5324909747292419,
|
| 420 |
+
"grad_norm": 0.3806765377521515,
|
| 421 |
+
"learning_rate": 3.949661830734172e-05,
|
| 422 |
+
"loss": 0.0224,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5415162454873647,
|
| 427 |
+
"grad_norm": 0.27796250581741333,
|
| 428 |
+
"learning_rate": 3.788607987366069e-05,
|
| 429 |
+
"loss": 0.0253,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5505415162454874,
|
| 434 |
+
"grad_norm": 0.24273499846458435,
|
| 435 |
+
"learning_rate": 3.628878783378302e-05,
|
| 436 |
+
"loss": 0.0208,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5595667870036101,
|
| 441 |
+
"grad_norm": 0.3810800611972809,
|
| 442 |
+
"learning_rate": 3.470648880310313e-05,
|
| 443 |
+
"loss": 0.0219,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5685920577617328,
|
| 448 |
+
"grad_norm": 0.31840479373931885,
|
| 449 |
+
"learning_rate": 3.3140913002379995e-05,
|
| 450 |
+
"loss": 0.0222,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5776173285198556,
|
| 455 |
+
"grad_norm": 0.2828656733036041,
|
| 456 |
+
"learning_rate": 3.1593772365766105e-05,
|
| 457 |
+
"loss": 0.0212,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5866425992779783,
|
| 462 |
+
"grad_norm": 0.22635580599308014,
|
| 463 |
+
"learning_rate": 3.006675866883275e-05,
|
| 464 |
+
"loss": 0.0213,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5956678700361011,
|
| 469 |
+
"grad_norm": 0.20679627358913422,
|
| 470 |
+
"learning_rate": 2.8561541678638142e-05,
|
| 471 |
+
"loss": 0.0239,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6046931407942239,
|
| 476 |
+
"grad_norm": 0.20941723883152008,
|
| 477 |
+
"learning_rate": 2.707976732786166e-05,
|
| 478 |
+
"loss": 0.0198,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6137184115523465,
|
| 483 |
+
"grad_norm": 0.23454327881336212,
|
| 484 |
+
"learning_rate": 2.562305591500069e-05,
|
| 485 |
+
"loss": 0.0199,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6227436823104693,
|
| 490 |
+
"grad_norm": 0.2705540060997009,
|
| 491 |
+
"learning_rate": 2.419300033259798e-05,
|
| 492 |
+
"loss": 0.0207,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.631768953068592,
|
| 497 |
+
"grad_norm": 0.2031233012676239,
|
| 498 |
+
"learning_rate": 2.279116432543705e-05,
|
| 499 |
+
"loss": 0.0195,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6407942238267148,
|
| 504 |
+
"grad_norm": 0.3306087553501129,
|
| 505 |
+
"learning_rate": 2.1419080780610123e-05,
|
| 506 |
+
"loss": 0.0237,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6498194945848376,
|
| 511 |
+
"grad_norm": 0.2521066963672638,
|
| 512 |
+
"learning_rate": 2.0078250051328784e-05,
|
| 513 |
+
"loss": 0.0201,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6588447653429603,
|
| 518 |
+
"grad_norm": 0.17102816700935364,
|
| 519 |
+
"learning_rate": 1.877013831630961e-05,
|
| 520 |
+
"loss": 0.0157,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6678700361010831,
|
| 525 |
+
"grad_norm": 0.2178596407175064,
|
| 526 |
+
"learning_rate": 1.749617597652934e-05,
|
| 527 |
+
"loss": 0.0212,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6768953068592057,
|
| 532 |
+
"grad_norm": 0.23762241005897522,
|
| 533 |
+
"learning_rate": 1.62577560911024e-05,
|
| 534 |
+
"loss": 0.0198,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6859205776173285,
|
| 539 |
+
"grad_norm": 0.24799777567386627,
|
| 540 |
+
"learning_rate": 1.5056232853991209e-05,
|
| 541 |
+
"loss": 0.0233,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6949458483754513,
|
| 546 |
+
"grad_norm": 0.1982533186674118,
|
| 547 |
+
"learning_rate": 1.389292011321498e-05,
|
| 548 |
+
"loss": 0.0177,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.703971119133574,
|
| 553 |
+
"grad_norm": 0.27399930357933044,
|
| 554 |
+
"learning_rate": 1.2769089934176126e-05,
|
| 555 |
+
"loss": 0.0189,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7129963898916968,
|
| 560 |
+
"grad_norm": 0.17227351665496826,
|
| 561 |
+
"learning_rate": 1.1685971208675539e-05,
|
| 562 |
+
"loss": 0.0184,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7220216606498195,
|
| 567 |
+
"grad_norm": 0.1276431679725647,
|
| 568 |
+
"learning_rate": 1.0644748311137376e-05,
|
| 569 |
+
"loss": 0.0183,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7310469314079422,
|
| 574 |
+
"grad_norm": 0.20766472816467285,
|
| 575 |
+
"learning_rate": 9.646559803512994e-06,
|
| 576 |
+
"loss": 0.019,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.740072202166065,
|
| 581 |
+
"grad_norm": 0.17726139724254608,
|
| 582 |
+
"learning_rate": 8.692497190280224e-06,
|
| 583 |
+
"loss": 0.0178,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7490974729241877,
|
| 588 |
+
"grad_norm": 0.2659638226032257,
|
| 589 |
+
"learning_rate": 7.783603724899257e-06,
|
| 590 |
+
"loss": 0.0186,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7581227436823105,
|
| 595 |
+
"grad_norm": 0.1689656376838684,
|
| 596 |
+
"learning_rate": 6.92087326903022e-06,
|
| 597 |
+
"loss": 0.0179,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7671480144404332,
|
| 602 |
+
"grad_norm": 0.2008303552865982,
|
| 603 |
+
"learning_rate": 6.1052492057601275e-06,
|
| 604 |
+
"loss": 0.0166,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.776173285198556,
|
| 609 |
+
"grad_norm": 0.2688104212284088,
|
| 610 |
+
"learning_rate": 5.337623408027293e-06,
|
| 611 |
+
"loss": 0.0201,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7851985559566786,
|
| 616 |
+
"grad_norm": 0.16218852996826172,
|
| 617 |
+
"learning_rate": 4.618835263371396e-06,
|
| 618 |
+
"loss": 0.0182,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7942238267148014,
|
| 623 |
+
"grad_norm": 0.19471345841884613,
|
| 624 |
+
"learning_rate": 3.949670756075447e-06,
|
| 625 |
+
"loss": 0.0178,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8032490974729242,
|
| 630 |
+
"grad_norm": 0.19546721875667572,
|
| 631 |
+
"learning_rate": 3.3308616077036115e-06,
|
| 632 |
+
"loss": 0.0178,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8122743682310469,
|
| 637 |
+
"grad_norm": 0.17014919221401215,
|
| 638 |
+
"learning_rate": 2.7630844769743757e-06,
|
| 639 |
+
"loss": 0.0152,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8212996389891697,
|
| 644 |
+
"grad_norm": 0.17462334036827087,
|
| 645 |
+
"learning_rate": 2.2469602198441573e-06,
|
| 646 |
+
"loss": 0.0159,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8303249097472925,
|
| 651 |
+
"grad_norm": 0.1929808259010315,
|
| 652 |
+
"learning_rate": 1.7830532106104747e-06,
|
| 653 |
+
"loss": 0.0152,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8393501805054152,
|
| 658 |
+
"grad_norm": 0.16477197408676147,
|
| 659 |
+
"learning_rate": 1.3718707247769135e-06,
|
| 660 |
+
"loss": 0.0159,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8483754512635379,
|
| 665 |
+
"grad_norm": 0.18839409947395325,
|
| 666 |
+
"learning_rate": 1.0138623843548078e-06,
|
| 667 |
+
"loss": 0.0143,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8574007220216606,
|
| 672 |
+
"grad_norm": 0.19429394602775574,
|
| 673 |
+
"learning_rate": 7.094196662081831e-07,
|
| 674 |
+
"loss": 0.0173,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8664259927797834,
|
| 679 |
+
"grad_norm": 0.15573933720588684,
|
| 680 |
+
"learning_rate": 4.5887547397955864e-07,
|
| 681 |
+
"loss": 0.0152,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8754512635379061,
|
| 686 |
+
"grad_norm": 0.16420625150203705,
|
| 687 |
+
"learning_rate": 2.625037740646763e-07,
|
| 688 |
+
"loss": 0.0181,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8844765342960289,
|
| 693 |
+
"grad_norm": 0.14045004546642303,
|
| 694 |
+
"learning_rate": 1.2051929603428825e-07,
|
| 695 |
+
"loss": 0.0163,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8935018050541517,
|
| 700 |
+
"grad_norm": 0.20185790956020355,
|
| 701 |
+
"learning_rate": 3.3077297830541584e-08,
|
| 702 |
+
"loss": 0.0189,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9025270758122743,
|
| 707 |
+
"grad_norm": 0.2897006571292877,
|
| 708 |
+
"learning_rate": 2.7339599464326627e-10,
|
| 709 |
+
"loss": 0.0205,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9025270758122743,
|
| 714 |
+
"step": 1000,
|
| 715 |
+
"total_flos": 0.0,
|
| 716 |
+
"train_loss": 0.04500286555290222,
|
| 717 |
+
"train_runtime": 685.5246,
|
| 718 |
+
"train_samples_per_second": 70.019,
|
| 719 |
+
"train_steps_per_second": 1.459
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 10,
|
| 723 |
+
"max_steps": 1000,
|
| 724 |
+
"num_input_tokens_seen": 0,
|
| 725 |
+
"num_train_epochs": 1,
|
| 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": 48,
|
| 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 sha256:783b80d81286a5b55670d0b48c8fb948472e733210763a68cc74c07e6c359f64
|
| 3 |
+
size 5304
|