Commit
•
1f402b7
1
Parent(s):
1c21ffe
Upload trainer_state.json
Browse files- trainer_state.json +452 -36
trainer_state.json
CHANGED
|
@@ -1,76 +1,492 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint":
|
| 4 |
-
"epoch": 4.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
-
"learning_rate": 1.
|
| 13 |
-
"loss": 1.
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"
|
| 19 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch":
|
| 24 |
-
"
|
| 25 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch":
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss":
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"
|
| 37 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"
|
| 43 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"
|
| 49 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch":
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch":
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss":
|
| 62 |
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch":
|
| 66 |
-
"
|
| 67 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
"step": 5000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
}
|
| 70 |
],
|
| 71 |
-
"max_steps":
|
| 72 |
"num_train_epochs": 5,
|
| 73 |
-
"total_flos":
|
| 74 |
"trial_name": null,
|
| 75 |
"trial_params": null
|
| 76 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.1290708780288696,
|
| 3 |
+
"best_model_checkpoint": "/home/ubuntu/ML/source_code/fine-tuned-legalroberta/checkpoint-17000",
|
| 4 |
+
"epoch": 4.779308405960078,
|
| 5 |
+
"global_step": 17000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 1.9437728422828227e-05,
|
| 13 |
+
"loss": 1.5399,
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.5552507638931274,
|
| 19 |
+
"eval_runtime": 2.1088,
|
| 20 |
+
"eval_samples_per_second": 72.078,
|
| 21 |
+
"eval_steps_per_second": 7.587,
|
| 22 |
+
"step": 500
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 1.8875456845656453e-05,
|
| 27 |
+
"loss": 1.3513,
|
| 28 |
"step": 1000
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 1.3741689920425415,
|
| 33 |
+
"eval_runtime": 2.1146,
|
| 34 |
+
"eval_samples_per_second": 71.881,
|
| 35 |
+
"eval_steps_per_second": 7.566,
|
| 36 |
+
"step": 1000
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.42,
|
| 40 |
+
"learning_rate": 1.831318526848468e-05,
|
| 41 |
+
"loss": 1.2809,
|
| 42 |
+
"step": 1500
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.42,
|
| 46 |
+
"eval_loss": 1.3449822664260864,
|
| 47 |
+
"eval_runtime": 2.1138,
|
| 48 |
+
"eval_samples_per_second": 71.909,
|
| 49 |
+
"eval_steps_per_second": 7.569,
|
| 50 |
"step": 1500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.56,
|
| 54 |
+
"learning_rate": 1.7750913691312905e-05,
|
| 55 |
+
"loss": 1.2484,
|
| 56 |
"step": 2000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.56,
|
| 60 |
+
"eval_loss": 1.3560587167739868,
|
| 61 |
+
"eval_runtime": 2.1078,
|
| 62 |
+
"eval_samples_per_second": 72.113,
|
| 63 |
+
"eval_steps_per_second": 7.591,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.7,
|
| 68 |
+
"learning_rate": 1.718864211414113e-05,
|
| 69 |
+
"loss": 1.2016,
|
| 70 |
"step": 2500
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"epoch": 0.7,
|
| 74 |
+
"eval_loss": 1.3128280639648438,
|
| 75 |
+
"eval_runtime": 2.1023,
|
| 76 |
+
"eval_samples_per_second": 72.303,
|
| 77 |
+
"eval_steps_per_second": 7.611,
|
| 78 |
+
"step": 2500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.84,
|
| 82 |
+
"learning_rate": 1.662637053696936e-05,
|
| 83 |
+
"loss": 1.1724,
|
| 84 |
"step": 3000
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.84,
|
| 88 |
+
"eval_loss": 1.3004732131958008,
|
| 89 |
+
"eval_runtime": 2.1125,
|
| 90 |
+
"eval_samples_per_second": 71.951,
|
| 91 |
+
"eval_steps_per_second": 7.574,
|
| 92 |
+
"step": 3000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.98,
|
| 96 |
+
"learning_rate": 1.606409895979758e-05,
|
| 97 |
+
"loss": 1.1664,
|
| 98 |
+
"step": 3500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.98,
|
| 102 |
+
"eval_loss": 1.2773096561431885,
|
| 103 |
+
"eval_runtime": 2.119,
|
| 104 |
+
"eval_samples_per_second": 71.732,
|
| 105 |
+
"eval_steps_per_second": 7.551,
|
| 106 |
"step": 3500
|
| 107 |
},
|
| 108 |
{
|
| 109 |
+
"epoch": 1.12,
|
| 110 |
+
"learning_rate": 1.550182738262581e-05,
|
| 111 |
+
"loss": 1.1399,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.12,
|
| 116 |
+
"eval_loss": 1.2295717000961304,
|
| 117 |
+
"eval_runtime": 2.119,
|
| 118 |
+
"eval_samples_per_second": 71.733,
|
| 119 |
+
"eval_steps_per_second": 7.551,
|
| 120 |
"step": 4000
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 1.27,
|
| 124 |
+
"learning_rate": 1.4939555805454037e-05,
|
| 125 |
+
"loss": 1.1136,
|
| 126 |
"step": 4500
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 1.27,
|
| 130 |
+
"eval_loss": 1.2814128398895264,
|
| 131 |
+
"eval_runtime": 2.1037,
|
| 132 |
+
"eval_samples_per_second": 72.255,
|
| 133 |
+
"eval_steps_per_second": 7.606,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.41,
|
| 138 |
+
"learning_rate": 1.4377284228282262e-05,
|
| 139 |
+
"loss": 1.1026,
|
| 140 |
"step": 5000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.41,
|
| 144 |
+
"eval_loss": 1.2653945684432983,
|
| 145 |
+
"eval_runtime": 2.1041,
|
| 146 |
+
"eval_samples_per_second": 72.24,
|
| 147 |
+
"eval_steps_per_second": 7.604,
|
| 148 |
+
"step": 5000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.55,
|
| 152 |
+
"learning_rate": 1.3815012651110486e-05,
|
| 153 |
+
"loss": 1.0943,
|
| 154 |
+
"step": 5500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.55,
|
| 158 |
+
"eval_loss": 1.210158109664917,
|
| 159 |
+
"eval_runtime": 2.1061,
|
| 160 |
+
"eval_samples_per_second": 72.17,
|
| 161 |
+
"eval_steps_per_second": 7.597,
|
| 162 |
+
"step": 5500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.69,
|
| 166 |
+
"learning_rate": 1.3252741073938714e-05,
|
| 167 |
+
"loss": 1.0768,
|
| 168 |
+
"step": 6000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.69,
|
| 172 |
+
"eval_loss": 1.1943004131317139,
|
| 173 |
+
"eval_runtime": 2.1055,
|
| 174 |
+
"eval_samples_per_second": 72.191,
|
| 175 |
+
"eval_steps_per_second": 7.599,
|
| 176 |
+
"step": 6000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.83,
|
| 180 |
+
"learning_rate": 1.269046949676694e-05,
|
| 181 |
+
"loss": 1.0741,
|
| 182 |
+
"step": 6500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.83,
|
| 186 |
+
"eval_loss": 1.2276432514190674,
|
| 187 |
+
"eval_runtime": 2.1096,
|
| 188 |
+
"eval_samples_per_second": 72.05,
|
| 189 |
+
"eval_steps_per_second": 7.584,
|
| 190 |
+
"step": 6500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.97,
|
| 194 |
+
"learning_rate": 1.2128197919595167e-05,
|
| 195 |
+
"loss": 1.0561,
|
| 196 |
+
"step": 7000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.97,
|
| 200 |
+
"eval_loss": 1.1935926675796509,
|
| 201 |
+
"eval_runtime": 2.1073,
|
| 202 |
+
"eval_samples_per_second": 72.129,
|
| 203 |
+
"eval_steps_per_second": 7.593,
|
| 204 |
+
"step": 7000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.11,
|
| 208 |
+
"learning_rate": 1.1565926342423393e-05,
|
| 209 |
+
"loss": 1.0384,
|
| 210 |
+
"step": 7500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.11,
|
| 214 |
+
"eval_loss": 1.2100528478622437,
|
| 215 |
+
"eval_runtime": 2.1142,
|
| 216 |
+
"eval_samples_per_second": 71.894,
|
| 217 |
+
"eval_steps_per_second": 7.568,
|
| 218 |
+
"step": 7500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.25,
|
| 222 |
+
"learning_rate": 1.1003654765251617e-05,
|
| 223 |
+
"loss": 1.037,
|
| 224 |
+
"step": 8000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.25,
|
| 228 |
+
"eval_loss": 1.1716400384902954,
|
| 229 |
+
"eval_runtime": 2.1168,
|
| 230 |
+
"eval_samples_per_second": 71.805,
|
| 231 |
+
"eval_steps_per_second": 7.558,
|
| 232 |
+
"step": 8000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.39,
|
| 236 |
+
"learning_rate": 1.0441383188079844e-05,
|
| 237 |
+
"loss": 1.0295,
|
| 238 |
+
"step": 8500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.39,
|
| 242 |
+
"eval_loss": 1.1552788019180298,
|
| 243 |
+
"eval_runtime": 2.114,
|
| 244 |
+
"eval_samples_per_second": 71.901,
|
| 245 |
+
"eval_steps_per_second": 7.569,
|
| 246 |
+
"step": 8500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.53,
|
| 250 |
+
"learning_rate": 9.87911161090807e-06,
|
| 251 |
+
"loss": 1.0255,
|
| 252 |
+
"step": 9000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.53,
|
| 256 |
+
"eval_loss": 1.211492896080017,
|
| 257 |
+
"eval_runtime": 2.1048,
|
| 258 |
+
"eval_samples_per_second": 72.217,
|
| 259 |
+
"eval_steps_per_second": 7.602,
|
| 260 |
+
"step": 9000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.67,
|
| 264 |
+
"learning_rate": 9.316840033736295e-06,
|
| 265 |
+
"loss": 1.0207,
|
| 266 |
+
"step": 9500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.67,
|
| 270 |
+
"eval_loss": 1.1855554580688477,
|
| 271 |
+
"eval_runtime": 2.1112,
|
| 272 |
+
"eval_samples_per_second": 71.998,
|
| 273 |
+
"eval_steps_per_second": 7.579,
|
| 274 |
+
"step": 9500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.81,
|
| 278 |
+
"learning_rate": 8.754568456564521e-06,
|
| 279 |
+
"loss": 1.0097,
|
| 280 |
+
"step": 10000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.81,
|
| 284 |
+
"eval_loss": 1.1690603494644165,
|
| 285 |
+
"eval_runtime": 2.1041,
|
| 286 |
+
"eval_samples_per_second": 72.238,
|
| 287 |
+
"eval_steps_per_second": 7.604,
|
| 288 |
+
"step": 10000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.95,
|
| 292 |
+
"learning_rate": 8.192296879392747e-06,
|
| 293 |
+
"loss": 0.9996,
|
| 294 |
+
"step": 10500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.95,
|
| 298 |
+
"eval_loss": 1.1681824922561646,
|
| 299 |
+
"eval_runtime": 2.107,
|
| 300 |
+
"eval_samples_per_second": 72.141,
|
| 301 |
+
"eval_steps_per_second": 7.594,
|
| 302 |
+
"step": 10500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.09,
|
| 306 |
+
"learning_rate": 7.630025302220974e-06,
|
| 307 |
+
"loss": 0.9912,
|
| 308 |
+
"step": 11000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.09,
|
| 312 |
+
"eval_loss": 1.200119972229004,
|
| 313 |
+
"eval_runtime": 2.1358,
|
| 314 |
+
"eval_samples_per_second": 71.168,
|
| 315 |
+
"eval_steps_per_second": 7.491,
|
| 316 |
+
"step": 11000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.23,
|
| 320 |
+
"learning_rate": 7.067753725049199e-06,
|
| 321 |
+
"loss": 0.9844,
|
| 322 |
+
"step": 11500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.23,
|
| 326 |
+
"eval_loss": 1.129595160484314,
|
| 327 |
+
"eval_runtime": 2.111,
|
| 328 |
+
"eval_samples_per_second": 72.004,
|
| 329 |
+
"eval_steps_per_second": 7.579,
|
| 330 |
+
"step": 11500
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.37,
|
| 334 |
+
"learning_rate": 6.505482147877426e-06,
|
| 335 |
+
"loss": 0.9908,
|
| 336 |
+
"step": 12000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.37,
|
| 340 |
+
"eval_loss": 1.1818418502807617,
|
| 341 |
+
"eval_runtime": 2.1165,
|
| 342 |
+
"eval_samples_per_second": 71.817,
|
| 343 |
+
"eval_steps_per_second": 7.56,
|
| 344 |
+
"step": 12000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.51,
|
| 348 |
+
"learning_rate": 5.943210570705651e-06,
|
| 349 |
+
"loss": 0.9798,
|
| 350 |
+
"step": 12500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.51,
|
| 354 |
+
"eval_loss": 1.1727079153060913,
|
| 355 |
+
"eval_runtime": 2.1056,
|
| 356 |
+
"eval_samples_per_second": 72.188,
|
| 357 |
+
"eval_steps_per_second": 7.599,
|
| 358 |
+
"step": 12500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.65,
|
| 362 |
+
"learning_rate": 5.380938993533877e-06,
|
| 363 |
+
"loss": 0.9699,
|
| 364 |
+
"step": 13000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.65,
|
| 368 |
+
"eval_loss": 1.2000970840454102,
|
| 369 |
+
"eval_runtime": 2.1003,
|
| 370 |
+
"eval_samples_per_second": 72.372,
|
| 371 |
+
"eval_steps_per_second": 7.618,
|
| 372 |
+
"step": 13000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.8,
|
| 376 |
+
"learning_rate": 4.818667416362104e-06,
|
| 377 |
+
"loss": 0.9706,
|
| 378 |
+
"step": 13500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.8,
|
| 382 |
+
"eval_loss": 1.1695332527160645,
|
| 383 |
+
"eval_runtime": 2.1004,
|
| 384 |
+
"eval_samples_per_second": 72.366,
|
| 385 |
+
"eval_steps_per_second": 7.617,
|
| 386 |
+
"step": 13500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.94,
|
| 390 |
+
"learning_rate": 4.2563958391903294e-06,
|
| 391 |
+
"loss": 0.974,
|
| 392 |
+
"step": 14000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.94,
|
| 396 |
+
"eval_loss": 1.1403844356536865,
|
| 397 |
+
"eval_runtime": 2.1018,
|
| 398 |
+
"eval_samples_per_second": 72.318,
|
| 399 |
+
"eval_steps_per_second": 7.612,
|
| 400 |
+
"step": 14000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.08,
|
| 404 |
+
"learning_rate": 3.694124262018555e-06,
|
| 405 |
+
"loss": 0.9561,
|
| 406 |
+
"step": 14500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.08,
|
| 410 |
+
"eval_loss": 1.1319295167922974,
|
| 411 |
+
"eval_runtime": 2.1093,
|
| 412 |
+
"eval_samples_per_second": 72.062,
|
| 413 |
+
"eval_steps_per_second": 7.586,
|
| 414 |
+
"step": 14500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.22,
|
| 418 |
+
"learning_rate": 3.1318526848467813e-06,
|
| 419 |
+
"loss": 0.9598,
|
| 420 |
+
"step": 15000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.22,
|
| 424 |
+
"eval_loss": 1.1552999019622803,
|
| 425 |
+
"eval_runtime": 2.116,
|
| 426 |
+
"eval_samples_per_second": 71.835,
|
| 427 |
+
"eval_steps_per_second": 7.562,
|
| 428 |
+
"step": 15000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.36,
|
| 432 |
+
"learning_rate": 2.569581107675007e-06,
|
| 433 |
+
"loss": 0.9626,
|
| 434 |
+
"step": 15500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.36,
|
| 438 |
+
"eval_loss": 1.1351696252822876,
|
| 439 |
+
"eval_runtime": 2.1126,
|
| 440 |
+
"eval_samples_per_second": 71.95,
|
| 441 |
+
"eval_steps_per_second": 7.574,
|
| 442 |
+
"step": 15500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.5,
|
| 446 |
+
"learning_rate": 2.007309530503233e-06,
|
| 447 |
+
"loss": 0.9631,
|
| 448 |
+
"step": 16000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.5,
|
| 452 |
+
"eval_loss": 1.181087851524353,
|
| 453 |
+
"eval_runtime": 2.1067,
|
| 454 |
+
"eval_samples_per_second": 72.151,
|
| 455 |
+
"eval_steps_per_second": 7.595,
|
| 456 |
+
"step": 16000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.64,
|
| 460 |
+
"learning_rate": 1.4450379533314593e-06,
|
| 461 |
+
"loss": 0.9485,
|
| 462 |
+
"step": 16500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.64,
|
| 466 |
+
"eval_loss": 1.1709717512130737,
|
| 467 |
+
"eval_runtime": 2.1077,
|
| 468 |
+
"eval_samples_per_second": 72.118,
|
| 469 |
+
"eval_steps_per_second": 7.591,
|
| 470 |
+
"step": 16500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.78,
|
| 474 |
+
"learning_rate": 8.827663761596852e-07,
|
| 475 |
+
"loss": 0.946,
|
| 476 |
+
"step": 17000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.78,
|
| 480 |
+
"eval_loss": 1.1290708780288696,
|
| 481 |
+
"eval_runtime": 2.1091,
|
| 482 |
+
"eval_samples_per_second": 72.067,
|
| 483 |
+
"eval_steps_per_second": 7.586,
|
| 484 |
+
"step": 17000
|
| 485 |
}
|
| 486 |
],
|
| 487 |
+
"max_steps": 17785,
|
| 488 |
"num_train_epochs": 5,
|
| 489 |
+
"total_flos": 4.47488103892009e+16,
|
| 490 |
"trial_name": null,
|
| 491 |
"trial_params": null
|
| 492 |
}
|