Commit
•
24db02e
1
Parent(s):
922952b
Upload trainer_state.json
Browse files- trainer_state.json +461 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,461 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0202020202020203,
|
| 5 |
+
"eval_steps": 20,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.050505050505050504,
|
| 13 |
+
"grad_norm": 0.1097809225320816,
|
| 14 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 15 |
+
"loss": 1.2164,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.10101010101010101,
|
| 20 |
+
"grad_norm": 0.16844922304153442,
|
| 21 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 22 |
+
"loss": 1.1774,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10101010101010101,
|
| 27 |
+
"eval_loss": 1.1594151258468628,
|
| 28 |
+
"eval_runtime": 295.5081,
|
| 29 |
+
"eval_samples_per_second": 2.382,
|
| 30 |
+
"eval_steps_per_second": 0.298,
|
| 31 |
+
"step": 20
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.15151515151515152,
|
| 35 |
+
"grad_norm": 0.3893352150917053,
|
| 36 |
+
"learning_rate": 8.999999999999999e-05,
|
| 37 |
+
"loss": 0.9942,
|
| 38 |
+
"step": 30
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.20202020202020202,
|
| 42 |
+
"grad_norm": 0.4363686740398407,
|
| 43 |
+
"learning_rate": 0.00011999999999999999,
|
| 44 |
+
"loss": 0.7259,
|
| 45 |
+
"step": 40
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.20202020202020202,
|
| 49 |
+
"eval_loss": 0.6028452515602112,
|
| 50 |
+
"eval_runtime": 294.8084,
|
| 51 |
+
"eval_samples_per_second": 2.388,
|
| 52 |
+
"eval_steps_per_second": 0.298,
|
| 53 |
+
"step": 40
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.25252525252525254,
|
| 57 |
+
"grad_norm": 0.40602898597717285,
|
| 58 |
+
"learning_rate": 0.00015,
|
| 59 |
+
"loss": 0.4454,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.30303030303030304,
|
| 64 |
+
"grad_norm": 0.2177702635526657,
|
| 65 |
+
"learning_rate": 0.00017999999999999998,
|
| 66 |
+
"loss": 0.5139,
|
| 67 |
+
"step": 60
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.30303030303030304,
|
| 71 |
+
"eval_loss": 0.3834875822067261,
|
| 72 |
+
"eval_runtime": 295.1108,
|
| 73 |
+
"eval_samples_per_second": 2.386,
|
| 74 |
+
"eval_steps_per_second": 0.298,
|
| 75 |
+
"step": 60
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.35353535353535354,
|
| 79 |
+
"grad_norm": 0.1777501404285431,
|
| 80 |
+
"learning_rate": 0.00020999999999999998,
|
| 81 |
+
"loss": 0.4096,
|
| 82 |
+
"step": 70
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.40404040404040403,
|
| 86 |
+
"grad_norm": 0.1749625951051712,
|
| 87 |
+
"learning_rate": 0.00023999999999999998,
|
| 88 |
+
"loss": 0.2589,
|
| 89 |
+
"step": 80
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.40404040404040403,
|
| 93 |
+
"eval_loss": 0.28596043586730957,
|
| 94 |
+
"eval_runtime": 294.9045,
|
| 95 |
+
"eval_samples_per_second": 2.387,
|
| 96 |
+
"eval_steps_per_second": 0.298,
|
| 97 |
+
"step": 80
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.45454545454545453,
|
| 101 |
+
"grad_norm": 0.2074197232723236,
|
| 102 |
+
"learning_rate": 0.00027,
|
| 103 |
+
"loss": 0.1932,
|
| 104 |
+
"step": 90
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.5050505050505051,
|
| 108 |
+
"grad_norm": 0.36931025981903076,
|
| 109 |
+
"learning_rate": 0.0003,
|
| 110 |
+
"loss": 0.2177,
|
| 111 |
+
"step": 100
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.5050505050505051,
|
| 115 |
+
"eval_loss": 0.25518086552619934,
|
| 116 |
+
"eval_runtime": 295.271,
|
| 117 |
+
"eval_samples_per_second": 2.384,
|
| 118 |
+
"eval_steps_per_second": 0.298,
|
| 119 |
+
"step": 100
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.5555555555555556,
|
| 123 |
+
"grad_norm": 0.20555801689624786,
|
| 124 |
+
"learning_rate": 0.00029,
|
| 125 |
+
"loss": 0.299,
|
| 126 |
+
"step": 110
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.6060606060606061,
|
| 130 |
+
"grad_norm": 0.14223843812942505,
|
| 131 |
+
"learning_rate": 0.00028,
|
| 132 |
+
"loss": 0.2508,
|
| 133 |
+
"step": 120
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.6060606060606061,
|
| 137 |
+
"eval_loss": 0.2106790542602539,
|
| 138 |
+
"eval_runtime": 294.889,
|
| 139 |
+
"eval_samples_per_second": 2.387,
|
| 140 |
+
"eval_steps_per_second": 0.298,
|
| 141 |
+
"step": 120
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.6565656565656566,
|
| 145 |
+
"grad_norm": 0.14103205502033234,
|
| 146 |
+
"learning_rate": 0.00027,
|
| 147 |
+
"loss": 0.1666,
|
| 148 |
+
"step": 130
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.7070707070707071,
|
| 152 |
+
"grad_norm": 0.1655016392469406,
|
| 153 |
+
"learning_rate": 0.00026,
|
| 154 |
+
"loss": 0.1511,
|
| 155 |
+
"step": 140
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.7070707070707071,
|
| 159 |
+
"eval_loss": 0.20343247056007385,
|
| 160 |
+
"eval_runtime": 294.7497,
|
| 161 |
+
"eval_samples_per_second": 2.388,
|
| 162 |
+
"eval_steps_per_second": 0.299,
|
| 163 |
+
"step": 140
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7575757575757576,
|
| 167 |
+
"grad_norm": 0.33550018072128296,
|
| 168 |
+
"learning_rate": 0.00025,
|
| 169 |
+
"loss": 0.1798,
|
| 170 |
+
"step": 150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8080808080808081,
|
| 174 |
+
"grad_norm": 0.18753309547901154,
|
| 175 |
+
"learning_rate": 0.00023999999999999998,
|
| 176 |
+
"loss": 0.245,
|
| 177 |
+
"step": 160
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8080808080808081,
|
| 181 |
+
"eval_loss": 0.19205082952976227,
|
| 182 |
+
"eval_runtime": 294.8122,
|
| 183 |
+
"eval_samples_per_second": 2.388,
|
| 184 |
+
"eval_steps_per_second": 0.298,
|
| 185 |
+
"step": 160
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.8585858585858586,
|
| 189 |
+
"grad_norm": 0.18931691348552704,
|
| 190 |
+
"learning_rate": 0.00023,
|
| 191 |
+
"loss": 0.2125,
|
| 192 |
+
"step": 170
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.9090909090909091,
|
| 196 |
+
"grad_norm": 0.1642046868801117,
|
| 197 |
+
"learning_rate": 0.00021999999999999995,
|
| 198 |
+
"loss": 0.1479,
|
| 199 |
+
"step": 180
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.9090909090909091,
|
| 203 |
+
"eval_loss": 0.1822979599237442,
|
| 204 |
+
"eval_runtime": 295.0975,
|
| 205 |
+
"eval_samples_per_second": 2.386,
|
| 206 |
+
"eval_steps_per_second": 0.298,
|
| 207 |
+
"step": 180
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.9595959595959596,
|
| 211 |
+
"grad_norm": 0.14441154897212982,
|
| 212 |
+
"learning_rate": 0.00020999999999999998,
|
| 213 |
+
"loss": 0.1399,
|
| 214 |
+
"step": 190
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.0101010101010102,
|
| 218 |
+
"grad_norm": 0.20781415700912476,
|
| 219 |
+
"learning_rate": 0.00019999999999999998,
|
| 220 |
+
"loss": 0.17,
|
| 221 |
+
"step": 200
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.0101010101010102,
|
| 225 |
+
"eval_loss": 0.18901540338993073,
|
| 226 |
+
"eval_runtime": 294.9368,
|
| 227 |
+
"eval_samples_per_second": 2.387,
|
| 228 |
+
"eval_steps_per_second": 0.298,
|
| 229 |
+
"step": 200
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.0606060606060606,
|
| 233 |
+
"grad_norm": 0.18098795413970947,
|
| 234 |
+
"learning_rate": 0.00018999999999999998,
|
| 235 |
+
"loss": 0.2102,
|
| 236 |
+
"step": 210
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.1111111111111112,
|
| 240 |
+
"grad_norm": 0.12826304137706757,
|
| 241 |
+
"learning_rate": 0.00017999999999999998,
|
| 242 |
+
"loss": 0.2085,
|
| 243 |
+
"step": 220
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.1111111111111112,
|
| 247 |
+
"eval_loss": 0.17017190158367157,
|
| 248 |
+
"eval_runtime": 295.0936,
|
| 249 |
+
"eval_samples_per_second": 2.386,
|
| 250 |
+
"eval_steps_per_second": 0.298,
|
| 251 |
+
"step": 220
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.1616161616161615,
|
| 255 |
+
"grad_norm": 0.13167431950569153,
|
| 256 |
+
"learning_rate": 0.00016999999999999999,
|
| 257 |
+
"loss": 0.1399,
|
| 258 |
+
"step": 230
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.2121212121212122,
|
| 262 |
+
"grad_norm": 0.13884004950523376,
|
| 263 |
+
"learning_rate": 0.00015999999999999999,
|
| 264 |
+
"loss": 0.1311,
|
| 265 |
+
"step": 240
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.2121212121212122,
|
| 269 |
+
"eval_loss": 0.16837279498577118,
|
| 270 |
+
"eval_runtime": 294.8077,
|
| 271 |
+
"eval_samples_per_second": 2.388,
|
| 272 |
+
"eval_steps_per_second": 0.298,
|
| 273 |
+
"step": 240
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.2626262626262625,
|
| 277 |
+
"grad_norm": 0.17478157579898834,
|
| 278 |
+
"learning_rate": 0.00015,
|
| 279 |
+
"loss": 0.1559,
|
| 280 |
+
"step": 250
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.3131313131313131,
|
| 284 |
+
"grad_norm": 0.16669002175331116,
|
| 285 |
+
"learning_rate": 0.00014,
|
| 286 |
+
"loss": 0.205,
|
| 287 |
+
"step": 260
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.3131313131313131,
|
| 291 |
+
"eval_loss": 0.16558928787708282,
|
| 292 |
+
"eval_runtime": 294.9081,
|
| 293 |
+
"eval_samples_per_second": 2.387,
|
| 294 |
+
"eval_steps_per_second": 0.298,
|
| 295 |
+
"step": 260
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.3636363636363638,
|
| 299 |
+
"grad_norm": 0.14852865040302277,
|
| 300 |
+
"learning_rate": 0.00013,
|
| 301 |
+
"loss": 0.1826,
|
| 302 |
+
"step": 270
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.4141414141414141,
|
| 306 |
+
"grad_norm": 0.14237315952777863,
|
| 307 |
+
"learning_rate": 0.00011999999999999999,
|
| 308 |
+
"loss": 0.1329,
|
| 309 |
+
"step": 280
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.4141414141414141,
|
| 313 |
+
"eval_loss": 0.15846213698387146,
|
| 314 |
+
"eval_runtime": 295.0174,
|
| 315 |
+
"eval_samples_per_second": 2.386,
|
| 316 |
+
"eval_steps_per_second": 0.298,
|
| 317 |
+
"step": 280
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.4646464646464645,
|
| 321 |
+
"grad_norm": 0.1401522010564804,
|
| 322 |
+
"learning_rate": 0.00010999999999999998,
|
| 323 |
+
"loss": 0.1205,
|
| 324 |
+
"step": 290
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.5151515151515151,
|
| 328 |
+
"grad_norm": 0.18983517587184906,
|
| 329 |
+
"learning_rate": 9.999999999999999e-05,
|
| 330 |
+
"loss": 0.1607,
|
| 331 |
+
"step": 300
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.5151515151515151,
|
| 335 |
+
"eval_loss": 0.16163967549800873,
|
| 336 |
+
"eval_runtime": 294.807,
|
| 337 |
+
"eval_samples_per_second": 2.388,
|
| 338 |
+
"eval_steps_per_second": 0.299,
|
| 339 |
+
"step": 300
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.5656565656565657,
|
| 343 |
+
"grad_norm": 0.1500634402036667,
|
| 344 |
+
"learning_rate": 8.999999999999999e-05,
|
| 345 |
+
"loss": 0.2121,
|
| 346 |
+
"step": 310
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.6161616161616161,
|
| 350 |
+
"grad_norm": 0.15094490349292755,
|
| 351 |
+
"learning_rate": 7.999999999999999e-05,
|
| 352 |
+
"loss": 0.1713,
|
| 353 |
+
"step": 320
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6161616161616161,
|
| 357 |
+
"eval_loss": 0.1540195345878601,
|
| 358 |
+
"eval_runtime": 295.157,
|
| 359 |
+
"eval_samples_per_second": 2.385,
|
| 360 |
+
"eval_steps_per_second": 0.298,
|
| 361 |
+
"step": 320
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.6666666666666665,
|
| 365 |
+
"grad_norm": 0.15443100035190582,
|
| 366 |
+
"learning_rate": 7e-05,
|
| 367 |
+
"loss": 0.129,
|
| 368 |
+
"step": 330
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.7171717171717171,
|
| 372 |
+
"grad_norm": 0.1568954586982727,
|
| 373 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 374 |
+
"loss": 0.1196,
|
| 375 |
+
"step": 340
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.7171717171717171,
|
| 379 |
+
"eval_loss": 0.15283828973770142,
|
| 380 |
+
"eval_runtime": 295.3671,
|
| 381 |
+
"eval_samples_per_second": 2.383,
|
| 382 |
+
"eval_steps_per_second": 0.298,
|
| 383 |
+
"step": 340
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.7676767676767677,
|
| 387 |
+
"grad_norm": 0.19448110461235046,
|
| 388 |
+
"learning_rate": 4.9999999999999996e-05,
|
| 389 |
+
"loss": 0.1603,
|
| 390 |
+
"step": 350
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.8181818181818183,
|
| 394 |
+
"grad_norm": 0.12574005126953125,
|
| 395 |
+
"learning_rate": 3.9999999999999996e-05,
|
| 396 |
+
"loss": 0.1722,
|
| 397 |
+
"step": 360
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.8181818181818183,
|
| 401 |
+
"eval_loss": 0.15110255777835846,
|
| 402 |
+
"eval_runtime": 295.1101,
|
| 403 |
+
"eval_samples_per_second": 2.386,
|
| 404 |
+
"eval_steps_per_second": 0.298,
|
| 405 |
+
"step": 360
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.8686868686868687,
|
| 409 |
+
"grad_norm": 0.11745467782020569,
|
| 410 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 411 |
+
"loss": 0.1669,
|
| 412 |
+
"step": 370
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.9191919191919191,
|
| 416 |
+
"grad_norm": 0.1326703280210495,
|
| 417 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 418 |
+
"loss": 0.1243,
|
| 419 |
+
"step": 380
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.9191919191919191,
|
| 423 |
+
"eval_loss": 0.1487378031015396,
|
| 424 |
+
"eval_runtime": 295.0041,
|
| 425 |
+
"eval_samples_per_second": 2.386,
|
| 426 |
+
"eval_steps_per_second": 0.298,
|
| 427 |
+
"step": 380
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.9696969696969697,
|
| 431 |
+
"grad_norm": 0.1454104483127594,
|
| 432 |
+
"learning_rate": 9.999999999999999e-06,
|
| 433 |
+
"loss": 0.1148,
|
| 434 |
+
"step": 390
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.0202020202020203,
|
| 438 |
+
"grad_norm": 0.1772913634777069,
|
| 439 |
+
"learning_rate": 0.0,
|
| 440 |
+
"loss": 0.1633,
|
| 441 |
+
"step": 400
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.0202020202020203,
|
| 445 |
+
"eval_loss": 0.14871937036514282,
|
| 446 |
+
"eval_runtime": 295.1345,
|
| 447 |
+
"eval_samples_per_second": 2.385,
|
| 448 |
+
"eval_steps_per_second": 0.298,
|
| 449 |
+
"step": 400
|
| 450 |
+
}
|
| 451 |
+
],
|
| 452 |
+
"logging_steps": 10,
|
| 453 |
+
"max_steps": 400,
|
| 454 |
+
"num_input_tokens_seen": 0,
|
| 455 |
+
"num_train_epochs": 3,
|
| 456 |
+
"save_steps": 20,
|
| 457 |
+
"total_flos": 2.2256168195948544e+17,
|
| 458 |
+
"train_batch_size": 8,
|
| 459 |
+
"trial_name": null,
|
| 460 |
+
"trial_params": null
|
| 461 |
+
}
|