Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +558 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,558 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.6794742345809937,
|
| 3 |
+
"best_model_checkpoint": "models/dehanalkautsar/mbert-uncased-modified_embedding_table-en/checkpoint-70000",
|
| 4 |
+
"epoch": 0.624464048155099,
|
| 5 |
+
"eval_steps": 2000,
|
| 6 |
+
"global_step": 70000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.017841829947288543,
|
| 13 |
+
"grad_norm": 31.71474266052246,
|
| 14 |
+
"learning_rate": 4.9977697687696264e-05,
|
| 15 |
+
"loss": 5.6683,
|
| 16 |
+
"step": 2000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.017841829947288543,
|
| 20 |
+
"eval_loss": 3.4221630096435547,
|
| 21 |
+
"eval_runtime": 73.4656,
|
| 22 |
+
"eval_samples_per_second": 136.118,
|
| 23 |
+
"eval_steps_per_second": 2.137,
|
| 24 |
+
"step": 2000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.035683659894577086,
|
| 28 |
+
"grad_norm": 20.9453182220459,
|
| 29 |
+
"learning_rate": 4.995539537539252e-05,
|
| 30 |
+
"loss": 3.1985,
|
| 31 |
+
"step": 4000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.035683659894577086,
|
| 35 |
+
"eval_loss": 2.7761876583099365,
|
| 36 |
+
"eval_runtime": 73.4155,
|
| 37 |
+
"eval_samples_per_second": 136.211,
|
| 38 |
+
"eval_steps_per_second": 2.139,
|
| 39 |
+
"step": 4000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.05352548984186563,
|
| 43 |
+
"grad_norm": 18.989194869995117,
|
| 44 |
+
"learning_rate": 4.993309306308878e-05,
|
| 45 |
+
"loss": 2.7816,
|
| 46 |
+
"step": 6000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.05352548984186563,
|
| 50 |
+
"eval_loss": 2.5367112159729004,
|
| 51 |
+
"eval_runtime": 73.3987,
|
| 52 |
+
"eval_samples_per_second": 136.242,
|
| 53 |
+
"eval_steps_per_second": 2.139,
|
| 54 |
+
"step": 6000
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.07136731978915417,
|
| 58 |
+
"grad_norm": 20.059371948242188,
|
| 59 |
+
"learning_rate": 4.991079075078504e-05,
|
| 60 |
+
"loss": 2.5767,
|
| 61 |
+
"step": 8000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.07136731978915417,
|
| 65 |
+
"eval_loss": 2.3765993118286133,
|
| 66 |
+
"eval_runtime": 73.3358,
|
| 67 |
+
"eval_samples_per_second": 136.359,
|
| 68 |
+
"eval_steps_per_second": 2.141,
|
| 69 |
+
"step": 8000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.08920914973644271,
|
| 73 |
+
"grad_norm": 18.088693618774414,
|
| 74 |
+
"learning_rate": 4.9888488438481305e-05,
|
| 75 |
+
"loss": 2.4472,
|
| 76 |
+
"step": 10000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.08920914973644271,
|
| 80 |
+
"eval_loss": 2.261122226715088,
|
| 81 |
+
"eval_runtime": 73.3532,
|
| 82 |
+
"eval_samples_per_second": 136.327,
|
| 83 |
+
"eval_steps_per_second": 2.14,
|
| 84 |
+
"step": 10000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.10705097968373126,
|
| 88 |
+
"grad_norm": 18.355438232421875,
|
| 89 |
+
"learning_rate": 4.986618612617757e-05,
|
| 90 |
+
"loss": 2.3517,
|
| 91 |
+
"step": 12000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.10705097968373126,
|
| 95 |
+
"eval_loss": 2.197890520095825,
|
| 96 |
+
"eval_runtime": 73.2175,
|
| 97 |
+
"eval_samples_per_second": 136.579,
|
| 98 |
+
"eval_steps_per_second": 2.144,
|
| 99 |
+
"step": 12000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.1248928096310198,
|
| 103 |
+
"grad_norm": 19.445158004760742,
|
| 104 |
+
"learning_rate": 4.984388381387383e-05,
|
| 105 |
+
"loss": 2.2819,
|
| 106 |
+
"step": 14000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.1248928096310198,
|
| 110 |
+
"eval_loss": 2.135493040084839,
|
| 111 |
+
"eval_runtime": 73.523,
|
| 112 |
+
"eval_samples_per_second": 136.012,
|
| 113 |
+
"eval_steps_per_second": 2.135,
|
| 114 |
+
"step": 14000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.14273463957830834,
|
| 118 |
+
"grad_norm": 16.74393081665039,
|
| 119 |
+
"learning_rate": 4.9821581501570084e-05,
|
| 120 |
+
"loss": 2.2253,
|
| 121 |
+
"step": 16000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.14273463957830834,
|
| 125 |
+
"eval_loss": 2.0676965713500977,
|
| 126 |
+
"eval_runtime": 73.4191,
|
| 127 |
+
"eval_samples_per_second": 136.204,
|
| 128 |
+
"eval_steps_per_second": 2.138,
|
| 129 |
+
"step": 16000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1605764695255969,
|
| 133 |
+
"grad_norm": 18.755170822143555,
|
| 134 |
+
"learning_rate": 4.9799279189266346e-05,
|
| 135 |
+
"loss": 2.1737,
|
| 136 |
+
"step": 18000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1605764695255969,
|
| 140 |
+
"eval_loss": 2.0221915245056152,
|
| 141 |
+
"eval_runtime": 73.3608,
|
| 142 |
+
"eval_samples_per_second": 136.313,
|
| 143 |
+
"eval_steps_per_second": 2.14,
|
| 144 |
+
"step": 18000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.17841829947288543,
|
| 148 |
+
"grad_norm": 18.238853454589844,
|
| 149 |
+
"learning_rate": 4.977697687696261e-05,
|
| 150 |
+
"loss": 2.1339,
|
| 151 |
+
"step": 20000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.17841829947288543,
|
| 155 |
+
"eval_loss": 1.9968066215515137,
|
| 156 |
+
"eval_runtime": 73.376,
|
| 157 |
+
"eval_samples_per_second": 136.284,
|
| 158 |
+
"eval_steps_per_second": 2.14,
|
| 159 |
+
"step": 20000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.19626012942017398,
|
| 163 |
+
"grad_norm": 16.624298095703125,
|
| 164 |
+
"learning_rate": 4.975467456465886e-05,
|
| 165 |
+
"loss": 2.097,
|
| 166 |
+
"step": 22000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.19626012942017398,
|
| 170 |
+
"eval_loss": 1.973600149154663,
|
| 171 |
+
"eval_runtime": 73.4156,
|
| 172 |
+
"eval_samples_per_second": 136.211,
|
| 173 |
+
"eval_steps_per_second": 2.139,
|
| 174 |
+
"step": 22000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.2141019593674625,
|
| 178 |
+
"grad_norm": 21.57083511352539,
|
| 179 |
+
"learning_rate": 4.9732372252355125e-05,
|
| 180 |
+
"loss": 2.0679,
|
| 181 |
+
"step": 24000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.2141019593674625,
|
| 185 |
+
"eval_loss": 1.9505703449249268,
|
| 186 |
+
"eval_runtime": 73.2384,
|
| 187 |
+
"eval_samples_per_second": 136.54,
|
| 188 |
+
"eval_steps_per_second": 2.144,
|
| 189 |
+
"step": 24000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.23194378931475107,
|
| 193 |
+
"grad_norm": 18.031625747680664,
|
| 194 |
+
"learning_rate": 4.971006994005139e-05,
|
| 195 |
+
"loss": 2.0474,
|
| 196 |
+
"step": 26000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.23194378931475107,
|
| 200 |
+
"eval_loss": 1.9203472137451172,
|
| 201 |
+
"eval_runtime": 73.2703,
|
| 202 |
+
"eval_samples_per_second": 136.481,
|
| 203 |
+
"eval_steps_per_second": 2.143,
|
| 204 |
+
"step": 26000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.2497856192620396,
|
| 208 |
+
"grad_norm": 18.387907028198242,
|
| 209 |
+
"learning_rate": 4.968776762774764e-05,
|
| 210 |
+
"loss": 2.0235,
|
| 211 |
+
"step": 28000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.2497856192620396,
|
| 215 |
+
"eval_loss": 1.9067487716674805,
|
| 216 |
+
"eval_runtime": 73.2711,
|
| 217 |
+
"eval_samples_per_second": 136.479,
|
| 218 |
+
"eval_steps_per_second": 2.143,
|
| 219 |
+
"step": 28000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2676274492093281,
|
| 223 |
+
"grad_norm": 16.74208641052246,
|
| 224 |
+
"learning_rate": 4.966546531544391e-05,
|
| 225 |
+
"loss": 2.0007,
|
| 226 |
+
"step": 30000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.2676274492093281,
|
| 230 |
+
"eval_loss": 1.8875941038131714,
|
| 231 |
+
"eval_runtime": 73.3071,
|
| 232 |
+
"eval_samples_per_second": 136.412,
|
| 233 |
+
"eval_steps_per_second": 2.142,
|
| 234 |
+
"step": 30000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2854692791566167,
|
| 238 |
+
"grad_norm": 17.28813934326172,
|
| 239 |
+
"learning_rate": 4.964316300314017e-05,
|
| 240 |
+
"loss": 1.9809,
|
| 241 |
+
"step": 32000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2854692791566167,
|
| 245 |
+
"eval_loss": 1.8658957481384277,
|
| 246 |
+
"eval_runtime": 73.2999,
|
| 247 |
+
"eval_samples_per_second": 136.426,
|
| 248 |
+
"eval_steps_per_second": 2.142,
|
| 249 |
+
"step": 32000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.30331110910390524,
|
| 253 |
+
"grad_norm": 17.0612735748291,
|
| 254 |
+
"learning_rate": 4.962086069083643e-05,
|
| 255 |
+
"loss": 1.9672,
|
| 256 |
+
"step": 34000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.30331110910390524,
|
| 260 |
+
"eval_loss": 1.8565300703048706,
|
| 261 |
+
"eval_runtime": 73.3279,
|
| 262 |
+
"eval_samples_per_second": 136.374,
|
| 263 |
+
"eval_steps_per_second": 2.141,
|
| 264 |
+
"step": 34000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.3211529390511938,
|
| 268 |
+
"grad_norm": 17.805253982543945,
|
| 269 |
+
"learning_rate": 4.959855837853269e-05,
|
| 270 |
+
"loss": 1.9517,
|
| 271 |
+
"step": 36000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.3211529390511938,
|
| 275 |
+
"eval_loss": 1.8271287679672241,
|
| 276 |
+
"eval_runtime": 73.2647,
|
| 277 |
+
"eval_samples_per_second": 136.491,
|
| 278 |
+
"eval_steps_per_second": 2.143,
|
| 279 |
+
"step": 36000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.33899476899848235,
|
| 283 |
+
"grad_norm": 16.978797912597656,
|
| 284 |
+
"learning_rate": 4.957625606622895e-05,
|
| 285 |
+
"loss": 1.9358,
|
| 286 |
+
"step": 38000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.33899476899848235,
|
| 290 |
+
"eval_loss": 1.8138540983200073,
|
| 291 |
+
"eval_runtime": 73.3003,
|
| 292 |
+
"eval_samples_per_second": 136.425,
|
| 293 |
+
"eval_steps_per_second": 2.142,
|
| 294 |
+
"step": 38000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.35683659894577086,
|
| 298 |
+
"grad_norm": 18.134506225585938,
|
| 299 |
+
"learning_rate": 4.955395375392521e-05,
|
| 300 |
+
"loss": 1.9204,
|
| 301 |
+
"step": 40000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.35683659894577086,
|
| 305 |
+
"eval_loss": 1.8061386346817017,
|
| 306 |
+
"eval_runtime": 73.3032,
|
| 307 |
+
"eval_samples_per_second": 136.42,
|
| 308 |
+
"eval_steps_per_second": 2.142,
|
| 309 |
+
"step": 40000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.3746784288930594,
|
| 313 |
+
"grad_norm": 16.018447875976562,
|
| 314 |
+
"learning_rate": 4.953165144162147e-05,
|
| 315 |
+
"loss": 1.9103,
|
| 316 |
+
"step": 42000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.3746784288930594,
|
| 320 |
+
"eval_loss": 1.790651559829712,
|
| 321 |
+
"eval_runtime": 73.3055,
|
| 322 |
+
"eval_samples_per_second": 136.415,
|
| 323 |
+
"eval_steps_per_second": 2.142,
|
| 324 |
+
"step": 42000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.39252025884034797,
|
| 328 |
+
"grad_norm": 18.30422592163086,
|
| 329 |
+
"learning_rate": 4.950934912931773e-05,
|
| 330 |
+
"loss": 1.8984,
|
| 331 |
+
"step": 44000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.39252025884034797,
|
| 335 |
+
"eval_loss": 1.787701964378357,
|
| 336 |
+
"eval_runtime": 73.3134,
|
| 337 |
+
"eval_samples_per_second": 136.401,
|
| 338 |
+
"eval_steps_per_second": 2.141,
|
| 339 |
+
"step": 44000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.4103620887876365,
|
| 343 |
+
"grad_norm": 16.60624122619629,
|
| 344 |
+
"learning_rate": 4.9487046817013986e-05,
|
| 345 |
+
"loss": 1.89,
|
| 346 |
+
"step": 46000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.4103620887876365,
|
| 350 |
+
"eval_loss": 1.7718769311904907,
|
| 351 |
+
"eval_runtime": 73.3399,
|
| 352 |
+
"eval_samples_per_second": 136.351,
|
| 353 |
+
"eval_steps_per_second": 2.141,
|
| 354 |
+
"step": 46000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.428203918734925,
|
| 358 |
+
"grad_norm": 15.059417724609375,
|
| 359 |
+
"learning_rate": 4.964316300314017e-05,
|
| 360 |
+
"loss": 1.8775,
|
| 361 |
+
"step": 48000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.428203918734925,
|
| 365 |
+
"eval_loss": 1.763095736503601,
|
| 366 |
+
"eval_runtime": 73.3521,
|
| 367 |
+
"eval_samples_per_second": 136.329,
|
| 368 |
+
"eval_steps_per_second": 2.14,
|
| 369 |
+
"step": 48000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.4460457486822136,
|
| 373 |
+
"grad_norm": 17.129064559936523,
|
| 374 |
+
"learning_rate": 4.9628294794937674e-05,
|
| 375 |
+
"loss": 1.8687,
|
| 376 |
+
"step": 50000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.4460457486822136,
|
| 380 |
+
"eval_loss": 1.743654727935791,
|
| 381 |
+
"eval_runtime": 73.2498,
|
| 382 |
+
"eval_samples_per_second": 136.519,
|
| 383 |
+
"eval_steps_per_second": 2.143,
|
| 384 |
+
"step": 50000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.46388757862950214,
|
| 388 |
+
"grad_norm": 16.349536895751953,
|
| 389 |
+
"learning_rate": 4.947285443645707e-05,
|
| 390 |
+
"loss": 1.8632,
|
| 391 |
+
"step": 52000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.46388757862950214,
|
| 395 |
+
"eval_loss": 1.743268609046936,
|
| 396 |
+
"eval_runtime": 73.3921,
|
| 397 |
+
"eval_samples_per_second": 136.254,
|
| 398 |
+
"eval_steps_per_second": 2.139,
|
| 399 |
+
"step": 52000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.4817294085767907,
|
| 403 |
+
"grad_norm": 16.42721939086914,
|
| 404 |
+
"learning_rate": 4.9452579607090025e-05,
|
| 405 |
+
"loss": 1.8494,
|
| 406 |
+
"step": 54000
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.4817294085767907,
|
| 410 |
+
"eval_loss": 1.7285025119781494,
|
| 411 |
+
"eval_runtime": 73.2209,
|
| 412 |
+
"eval_samples_per_second": 136.573,
|
| 413 |
+
"eval_steps_per_second": 2.144,
|
| 414 |
+
"step": 54000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.4995712385240792,
|
| 418 |
+
"grad_norm": 17.674468994140625,
|
| 419 |
+
"learning_rate": 4.9432304777722994e-05,
|
| 420 |
+
"loss": 1.8404,
|
| 421 |
+
"step": 56000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.4995712385240792,
|
| 425 |
+
"eval_loss": 1.7261757850646973,
|
| 426 |
+
"eval_runtime": 73.1858,
|
| 427 |
+
"eval_samples_per_second": 136.639,
|
| 428 |
+
"eval_steps_per_second": 2.145,
|
| 429 |
+
"step": 56000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5174130684713678,
|
| 433 |
+
"grad_norm": 16.304468154907227,
|
| 434 |
+
"learning_rate": 4.941202994835596e-05,
|
| 435 |
+
"loss": 1.8308,
|
| 436 |
+
"step": 58000
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5174130684713678,
|
| 440 |
+
"eval_loss": 1.7157503366470337,
|
| 441 |
+
"eval_runtime": 73.1684,
|
| 442 |
+
"eval_samples_per_second": 136.671,
|
| 443 |
+
"eval_steps_per_second": 2.146,
|
| 444 |
+
"step": 58000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5352548984186563,
|
| 448 |
+
"grad_norm": 17.134702682495117,
|
| 449 |
+
"learning_rate": 4.939175511898892e-05,
|
| 450 |
+
"loss": 1.8245,
|
| 451 |
+
"step": 60000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5352548984186563,
|
| 455 |
+
"eval_loss": 1.7094610929489136,
|
| 456 |
+
"eval_runtime": 73.3218,
|
| 457 |
+
"eval_samples_per_second": 136.385,
|
| 458 |
+
"eval_steps_per_second": 2.141,
|
| 459 |
+
"step": 60000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.5530967283659448,
|
| 463 |
+
"grad_norm": 17.70859146118164,
|
| 464 |
+
"learning_rate": 4.937148028962189e-05,
|
| 465 |
+
"loss": 1.8201,
|
| 466 |
+
"step": 62000
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.5530967283659448,
|
| 470 |
+
"eval_loss": 1.70658540725708,
|
| 471 |
+
"eval_runtime": 73.3241,
|
| 472 |
+
"eval_samples_per_second": 136.381,
|
| 473 |
+
"eval_steps_per_second": 2.141,
|
| 474 |
+
"step": 62000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.5709385583132334,
|
| 478 |
+
"grad_norm": 16.962129592895508,
|
| 479 |
+
"learning_rate": 4.935120546025485e-05,
|
| 480 |
+
"loss": 1.8107,
|
| 481 |
+
"step": 64000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.5709385583132334,
|
| 485 |
+
"eval_loss": 1.6914931535720825,
|
| 486 |
+
"eval_runtime": 73.2479,
|
| 487 |
+
"eval_samples_per_second": 136.523,
|
| 488 |
+
"eval_steps_per_second": 2.143,
|
| 489 |
+
"step": 64000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.5887803882605219,
|
| 493 |
+
"grad_norm": 16.842283248901367,
|
| 494 |
+
"learning_rate": 4.933093063088781e-05,
|
| 495 |
+
"loss": 1.8027,
|
| 496 |
+
"step": 66000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.5887803882605219,
|
| 500 |
+
"eval_loss": 1.683428168296814,
|
| 501 |
+
"eval_runtime": 73.2102,
|
| 502 |
+
"eval_samples_per_second": 136.593,
|
| 503 |
+
"eval_steps_per_second": 2.145,
|
| 504 |
+
"step": 66000
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.6066222182078105,
|
| 508 |
+
"grad_norm": 17.075162887573242,
|
| 509 |
+
"learning_rate": 4.9310655801520775e-05,
|
| 510 |
+
"loss": 1.7991,
|
| 511 |
+
"step": 68000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.6066222182078105,
|
| 515 |
+
"eval_loss": 1.688643217086792,
|
| 516 |
+
"eval_runtime": 73.3407,
|
| 517 |
+
"eval_samples_per_second": 136.35,
|
| 518 |
+
"eval_steps_per_second": 2.141,
|
| 519 |
+
"step": 68000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.624464048155099,
|
| 523 |
+
"grad_norm": 17.02593231201172,
|
| 524 |
+
"learning_rate": 4.929038097215374e-05,
|
| 525 |
+
"loss": 1.7906,
|
| 526 |
+
"step": 70000
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.624464048155099,
|
| 530 |
+
"eval_loss": 1.6794742345809937,
|
| 531 |
+
"eval_runtime": 73.266,
|
| 532 |
+
"eval_samples_per_second": 136.489,
|
| 533 |
+
"eval_steps_per_second": 2.143,
|
| 534 |
+
"step": 70000
|
| 535 |
+
}
|
| 536 |
+
],
|
| 537 |
+
"logging_steps": 2000,
|
| 538 |
+
"max_steps": 4932224,
|
| 539 |
+
"num_input_tokens_seen": 0,
|
| 540 |
+
"num_train_epochs": 44,
|
| 541 |
+
"save_steps": 2000,
|
| 542 |
+
"stateful_callbacks": {
|
| 543 |
+
"TrainerControl": {
|
| 544 |
+
"args": {
|
| 545 |
+
"should_epoch_stop": false,
|
| 546 |
+
"should_evaluate": false,
|
| 547 |
+
"should_log": false,
|
| 548 |
+
"should_save": true,
|
| 549 |
+
"should_training_stop": false
|
| 550 |
+
},
|
| 551 |
+
"attributes": {}
|
| 552 |
+
}
|
| 553 |
+
},
|
| 554 |
+
"total_flos": 4.716630245376e+18,
|
| 555 |
+
"train_batch_size": 32,
|
| 556 |
+
"trial_name": null,
|
| 557 |
+
"trial_params": null
|
| 558 |
+
}
|