Upload trainer_state.json
Browse files- trainer_state.json +576 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,576 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5370857725978839,
|
| 5 |
+
"global_step": 20000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 1.973145711370106e-05,
|
| 13 |
+
"loss": 0.2809,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"eval_loss": 0.9594895839691162,
|
| 19 |
+
"eval_runtime": 2.5122,
|
| 20 |
+
"eval_samples_per_second": 1378.065,
|
| 21 |
+
"eval_steps_per_second": 7.563,
|
| 22 |
+
"step": 500
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.03,
|
| 26 |
+
"learning_rate": 1.946291422740212e-05,
|
| 27 |
+
"loss": 0.2549,
|
| 28 |
+
"step": 1000
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.03,
|
| 32 |
+
"eval_loss": 1.03871750831604,
|
| 33 |
+
"eval_runtime": 2.5032,
|
| 34 |
+
"eval_samples_per_second": 1383.049,
|
| 35 |
+
"eval_steps_per_second": 7.59,
|
| 36 |
+
"step": 1000
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.04,
|
| 40 |
+
"learning_rate": 1.9194371341103174e-05,
|
| 41 |
+
"loss": 0.2473,
|
| 42 |
+
"step": 1500
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.04,
|
| 46 |
+
"eval_loss": 0.969585120677948,
|
| 47 |
+
"eval_runtime": 2.5286,
|
| 48 |
+
"eval_samples_per_second": 1369.122,
|
| 49 |
+
"eval_steps_per_second": 7.514,
|
| 50 |
+
"step": 1500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.05,
|
| 54 |
+
"learning_rate": 1.8925828454804235e-05,
|
| 55 |
+
"loss": 0.2514,
|
| 56 |
+
"step": 2000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.05,
|
| 60 |
+
"eval_loss": 0.8840947151184082,
|
| 61 |
+
"eval_runtime": 2.5014,
|
| 62 |
+
"eval_samples_per_second": 1384.048,
|
| 63 |
+
"eval_steps_per_second": 7.596,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.07,
|
| 68 |
+
"learning_rate": 1.8657285568505293e-05,
|
| 69 |
+
"loss": 0.2481,
|
| 70 |
+
"step": 2500
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.07,
|
| 74 |
+
"eval_loss": 0.9262232780456543,
|
| 75 |
+
"eval_runtime": 2.509,
|
| 76 |
+
"eval_samples_per_second": 1379.819,
|
| 77 |
+
"eval_steps_per_second": 7.573,
|
| 78 |
+
"step": 2500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.08,
|
| 82 |
+
"learning_rate": 1.838874268220635e-05,
|
| 83 |
+
"loss": 0.2428,
|
| 84 |
+
"step": 3000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.08,
|
| 88 |
+
"eval_loss": 0.9006544947624207,
|
| 89 |
+
"eval_runtime": 2.5007,
|
| 90 |
+
"eval_samples_per_second": 1384.421,
|
| 91 |
+
"eval_steps_per_second": 7.598,
|
| 92 |
+
"step": 3000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.09,
|
| 96 |
+
"learning_rate": 1.8120199795907408e-05,
|
| 97 |
+
"loss": 0.2411,
|
| 98 |
+
"step": 3500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.09,
|
| 102 |
+
"eval_loss": 0.9295358657836914,
|
| 103 |
+
"eval_runtime": 2.5648,
|
| 104 |
+
"eval_samples_per_second": 1349.806,
|
| 105 |
+
"eval_steps_per_second": 7.408,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.11,
|
| 110 |
+
"learning_rate": 1.7851656909608465e-05,
|
| 111 |
+
"loss": 0.2393,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.11,
|
| 116 |
+
"eval_loss": 0.8647714853286743,
|
| 117 |
+
"eval_runtime": 2.5012,
|
| 118 |
+
"eval_samples_per_second": 1384.115,
|
| 119 |
+
"eval_steps_per_second": 7.596,
|
| 120 |
+
"step": 4000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.12,
|
| 124 |
+
"learning_rate": 1.7583114023309523e-05,
|
| 125 |
+
"loss": 0.2389,
|
| 126 |
+
"step": 4500
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.12,
|
| 130 |
+
"eval_loss": 0.8838908672332764,
|
| 131 |
+
"eval_runtime": 2.5044,
|
| 132 |
+
"eval_samples_per_second": 1382.376,
|
| 133 |
+
"eval_steps_per_second": 7.587,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.13,
|
| 138 |
+
"learning_rate": 1.7314571137010584e-05,
|
| 139 |
+
"loss": 0.2402,
|
| 140 |
+
"step": 5000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.13,
|
| 144 |
+
"eval_loss": 0.927254319190979,
|
| 145 |
+
"eval_runtime": 2.5033,
|
| 146 |
+
"eval_samples_per_second": 1382.963,
|
| 147 |
+
"eval_steps_per_second": 7.59,
|
| 148 |
+
"step": 5000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.15,
|
| 152 |
+
"learning_rate": 1.7046028250711638e-05,
|
| 153 |
+
"loss": 0.2342,
|
| 154 |
+
"step": 5500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.15,
|
| 158 |
+
"eval_loss": 0.828113317489624,
|
| 159 |
+
"eval_runtime": 2.503,
|
| 160 |
+
"eval_samples_per_second": 1383.137,
|
| 161 |
+
"eval_steps_per_second": 7.591,
|
| 162 |
+
"step": 5500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.16,
|
| 166 |
+
"learning_rate": 1.67774853644127e-05,
|
| 167 |
+
"loss": 0.2336,
|
| 168 |
+
"step": 6000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.16,
|
| 172 |
+
"eval_loss": 0.9690735936164856,
|
| 173 |
+
"eval_runtime": 2.5071,
|
| 174 |
+
"eval_samples_per_second": 1380.884,
|
| 175 |
+
"eval_steps_per_second": 7.579,
|
| 176 |
+
"step": 6000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.17,
|
| 180 |
+
"learning_rate": 1.6508942478113756e-05,
|
| 181 |
+
"loss": 0.2363,
|
| 182 |
+
"step": 6500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.17,
|
| 186 |
+
"eval_loss": 0.8396281003952026,
|
| 187 |
+
"eval_runtime": 2.5067,
|
| 188 |
+
"eval_samples_per_second": 1381.116,
|
| 189 |
+
"eval_steps_per_second": 7.58,
|
| 190 |
+
"step": 6500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.19,
|
| 194 |
+
"learning_rate": 1.6240399591814814e-05,
|
| 195 |
+
"loss": 0.2337,
|
| 196 |
+
"step": 7000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.19,
|
| 200 |
+
"eval_loss": 0.8618175387382507,
|
| 201 |
+
"eval_runtime": 2.5018,
|
| 202 |
+
"eval_samples_per_second": 1383.782,
|
| 203 |
+
"eval_steps_per_second": 7.594,
|
| 204 |
+
"step": 7000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.2,
|
| 208 |
+
"learning_rate": 1.597185670551587e-05,
|
| 209 |
+
"loss": 0.2307,
|
| 210 |
+
"step": 7500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.2,
|
| 214 |
+
"eval_loss": 0.9263376593589783,
|
| 215 |
+
"eval_runtime": 2.5033,
|
| 216 |
+
"eval_samples_per_second": 1382.991,
|
| 217 |
+
"eval_steps_per_second": 7.59,
|
| 218 |
+
"step": 7500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.21,
|
| 222 |
+
"learning_rate": 1.570331381921693e-05,
|
| 223 |
+
"loss": 0.2347,
|
| 224 |
+
"step": 8000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.21,
|
| 228 |
+
"eval_loss": 0.8525048494338989,
|
| 229 |
+
"eval_runtime": 2.5022,
|
| 230 |
+
"eval_samples_per_second": 1383.595,
|
| 231 |
+
"eval_steps_per_second": 7.593,
|
| 232 |
+
"step": 8000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.23,
|
| 236 |
+
"learning_rate": 1.543477093291799e-05,
|
| 237 |
+
"loss": 0.2294,
|
| 238 |
+
"step": 8500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.23,
|
| 242 |
+
"eval_loss": 0.9026502966880798,
|
| 243 |
+
"eval_runtime": 2.503,
|
| 244 |
+
"eval_samples_per_second": 1383.13,
|
| 245 |
+
"eval_steps_per_second": 7.591,
|
| 246 |
+
"step": 8500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.24,
|
| 250 |
+
"learning_rate": 1.5166228046619047e-05,
|
| 251 |
+
"loss": 0.2297,
|
| 252 |
+
"step": 9000
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.24,
|
| 256 |
+
"eval_loss": 0.8343677520751953,
|
| 257 |
+
"eval_runtime": 2.503,
|
| 258 |
+
"eval_samples_per_second": 1383.136,
|
| 259 |
+
"eval_steps_per_second": 7.591,
|
| 260 |
+
"step": 9000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.26,
|
| 264 |
+
"learning_rate": 1.4897685160320103e-05,
|
| 265 |
+
"loss": 0.2294,
|
| 266 |
+
"step": 9500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.26,
|
| 270 |
+
"eval_loss": 0.8918474912643433,
|
| 271 |
+
"eval_runtime": 2.5026,
|
| 272 |
+
"eval_samples_per_second": 1383.364,
|
| 273 |
+
"eval_steps_per_second": 7.592,
|
| 274 |
+
"step": 9500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.27,
|
| 278 |
+
"learning_rate": 1.4629142274021162e-05,
|
| 279 |
+
"loss": 0.2322,
|
| 280 |
+
"step": 10000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.27,
|
| 284 |
+
"eval_loss": 0.832206130027771,
|
| 285 |
+
"eval_runtime": 2.5024,
|
| 286 |
+
"eval_samples_per_second": 1383.452,
|
| 287 |
+
"eval_steps_per_second": 7.593,
|
| 288 |
+
"step": 10000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.28,
|
| 292 |
+
"learning_rate": 1.436059938772222e-05,
|
| 293 |
+
"loss": 0.2327,
|
| 294 |
+
"step": 10500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.28,
|
| 298 |
+
"eval_loss": 0.8489925861358643,
|
| 299 |
+
"eval_runtime": 2.5027,
|
| 300 |
+
"eval_samples_per_second": 1383.288,
|
| 301 |
+
"eval_steps_per_second": 7.592,
|
| 302 |
+
"step": 10500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.3,
|
| 306 |
+
"learning_rate": 1.4092056501423279e-05,
|
| 307 |
+
"loss": 0.2313,
|
| 308 |
+
"step": 11000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.3,
|
| 312 |
+
"eval_loss": 0.8496459722518921,
|
| 313 |
+
"eval_runtime": 2.4999,
|
| 314 |
+
"eval_samples_per_second": 1384.839,
|
| 315 |
+
"eval_steps_per_second": 7.6,
|
| 316 |
+
"step": 11000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.31,
|
| 320 |
+
"learning_rate": 1.3823513615124335e-05,
|
| 321 |
+
"loss": 0.2305,
|
| 322 |
+
"step": 11500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.31,
|
| 326 |
+
"eval_loss": 0.9429409503936768,
|
| 327 |
+
"eval_runtime": 2.5651,
|
| 328 |
+
"eval_samples_per_second": 1349.664,
|
| 329 |
+
"eval_steps_per_second": 7.407,
|
| 330 |
+
"step": 11500
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.32,
|
| 334 |
+
"learning_rate": 1.3554970728825394e-05,
|
| 335 |
+
"loss": 0.2304,
|
| 336 |
+
"step": 12000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.32,
|
| 340 |
+
"eval_loss": 0.9687642455101013,
|
| 341 |
+
"eval_runtime": 2.497,
|
| 342 |
+
"eval_samples_per_second": 1386.456,
|
| 343 |
+
"eval_steps_per_second": 7.609,
|
| 344 |
+
"step": 12000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.34,
|
| 348 |
+
"learning_rate": 1.3286427842526453e-05,
|
| 349 |
+
"loss": 0.2295,
|
| 350 |
+
"step": 12500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.34,
|
| 354 |
+
"eval_loss": 0.9534400701522827,
|
| 355 |
+
"eval_runtime": 2.5012,
|
| 356 |
+
"eval_samples_per_second": 1384.153,
|
| 357 |
+
"eval_steps_per_second": 7.596,
|
| 358 |
+
"step": 12500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.35,
|
| 362 |
+
"learning_rate": 1.3017884956227511e-05,
|
| 363 |
+
"loss": 0.2328,
|
| 364 |
+
"step": 13000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.35,
|
| 368 |
+
"eval_loss": 0.9109036326408386,
|
| 369 |
+
"eval_runtime": 2.5038,
|
| 370 |
+
"eval_samples_per_second": 1382.713,
|
| 371 |
+
"eval_steps_per_second": 7.589,
|
| 372 |
+
"step": 13000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.36,
|
| 376 |
+
"learning_rate": 1.274934206992857e-05,
|
| 377 |
+
"loss": 0.2269,
|
| 378 |
+
"step": 13500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.36,
|
| 382 |
+
"eval_loss": 0.9277821779251099,
|
| 383 |
+
"eval_runtime": 2.5098,
|
| 384 |
+
"eval_samples_per_second": 1379.412,
|
| 385 |
+
"eval_steps_per_second": 7.57,
|
| 386 |
+
"step": 13500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.38,
|
| 390 |
+
"learning_rate": 1.2480799183629626e-05,
|
| 391 |
+
"loss": 0.2247,
|
| 392 |
+
"step": 14000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.38,
|
| 396 |
+
"eval_loss": 0.8515273332595825,
|
| 397 |
+
"eval_runtime": 2.5018,
|
| 398 |
+
"eval_samples_per_second": 1383.826,
|
| 399 |
+
"eval_steps_per_second": 7.595,
|
| 400 |
+
"step": 14000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.39,
|
| 404 |
+
"learning_rate": 1.2212256297330685e-05,
|
| 405 |
+
"loss": 0.2233,
|
| 406 |
+
"step": 14500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.39,
|
| 410 |
+
"eval_loss": 0.8787582516670227,
|
| 411 |
+
"eval_runtime": 2.5026,
|
| 412 |
+
"eval_samples_per_second": 1383.344,
|
| 413 |
+
"eval_steps_per_second": 7.592,
|
| 414 |
+
"step": 14500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.4,
|
| 418 |
+
"learning_rate": 1.1943713411031743e-05,
|
| 419 |
+
"loss": 0.2257,
|
| 420 |
+
"step": 15000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.4,
|
| 424 |
+
"eval_loss": 0.9345625638961792,
|
| 425 |
+
"eval_runtime": 2.502,
|
| 426 |
+
"eval_samples_per_second": 1383.68,
|
| 427 |
+
"eval_steps_per_second": 7.594,
|
| 428 |
+
"step": 15000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.42,
|
| 432 |
+
"learning_rate": 1.1675170524732802e-05,
|
| 433 |
+
"loss": 0.2263,
|
| 434 |
+
"step": 15500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.42,
|
| 438 |
+
"eval_loss": 0.9080494046211243,
|
| 439 |
+
"eval_runtime": 2.5009,
|
| 440 |
+
"eval_samples_per_second": 1384.285,
|
| 441 |
+
"eval_steps_per_second": 7.597,
|
| 442 |
+
"step": 15500
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.43,
|
| 446 |
+
"learning_rate": 1.1406627638433858e-05,
|
| 447 |
+
"loss": 0.2254,
|
| 448 |
+
"step": 16000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.43,
|
| 452 |
+
"eval_loss": 0.9714885950088501,
|
| 453 |
+
"eval_runtime": 2.5027,
|
| 454 |
+
"eval_samples_per_second": 1383.281,
|
| 455 |
+
"eval_steps_per_second": 7.592,
|
| 456 |
+
"step": 16000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.44,
|
| 460 |
+
"learning_rate": 1.1138084752134917e-05,
|
| 461 |
+
"loss": 0.2228,
|
| 462 |
+
"step": 16500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.44,
|
| 466 |
+
"eval_loss": 0.9011908769607544,
|
| 467 |
+
"eval_runtime": 2.4967,
|
| 468 |
+
"eval_samples_per_second": 1386.629,
|
| 469 |
+
"eval_steps_per_second": 7.61,
|
| 470 |
+
"step": 16500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.46,
|
| 474 |
+
"learning_rate": 1.0869541865835975e-05,
|
| 475 |
+
"loss": 0.2234,
|
| 476 |
+
"step": 17000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.46,
|
| 480 |
+
"eval_loss": 0.8457213044166565,
|
| 481 |
+
"eval_runtime": 2.5001,
|
| 482 |
+
"eval_samples_per_second": 1384.748,
|
| 483 |
+
"eval_steps_per_second": 7.6,
|
| 484 |
+
"step": 17000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.47,
|
| 488 |
+
"learning_rate": 1.0600998979537034e-05,
|
| 489 |
+
"loss": 0.221,
|
| 490 |
+
"step": 17500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.47,
|
| 494 |
+
"eval_loss": 0.9264540672302246,
|
| 495 |
+
"eval_runtime": 2.4947,
|
| 496 |
+
"eval_samples_per_second": 1387.758,
|
| 497 |
+
"eval_steps_per_second": 7.616,
|
| 498 |
+
"step": 17500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.48,
|
| 502 |
+
"learning_rate": 1.033245609323809e-05,
|
| 503 |
+
"loss": 0.2229,
|
| 504 |
+
"step": 18000
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.48,
|
| 508 |
+
"eval_loss": 0.9298267364501953,
|
| 509 |
+
"eval_runtime": 2.4937,
|
| 510 |
+
"eval_samples_per_second": 1388.279,
|
| 511 |
+
"eval_steps_per_second": 7.619,
|
| 512 |
+
"step": 18000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.5,
|
| 516 |
+
"learning_rate": 1.0063913206939149e-05,
|
| 517 |
+
"loss": 0.2245,
|
| 518 |
+
"step": 18500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.5,
|
| 522 |
+
"eval_loss": 0.9173328280448914,
|
| 523 |
+
"eval_runtime": 2.5,
|
| 524 |
+
"eval_samples_per_second": 1384.796,
|
| 525 |
+
"eval_steps_per_second": 7.6,
|
| 526 |
+
"step": 18500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.51,
|
| 530 |
+
"learning_rate": 9.795370320640208e-06,
|
| 531 |
+
"loss": 0.2232,
|
| 532 |
+
"step": 19000
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.51,
|
| 536 |
+
"eval_loss": 0.9208750128746033,
|
| 537 |
+
"eval_runtime": 2.4954,
|
| 538 |
+
"eval_samples_per_second": 1387.373,
|
| 539 |
+
"eval_steps_per_second": 7.614,
|
| 540 |
+
"step": 19000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.52,
|
| 544 |
+
"learning_rate": 9.526827434341266e-06,
|
| 545 |
+
"loss": 0.2224,
|
| 546 |
+
"step": 19500
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.52,
|
| 550 |
+
"eval_loss": 0.9066223502159119,
|
| 551 |
+
"eval_runtime": 2.502,
|
| 552 |
+
"eval_samples_per_second": 1383.684,
|
| 553 |
+
"eval_steps_per_second": 7.594,
|
| 554 |
+
"step": 19500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.54,
|
| 558 |
+
"learning_rate": 9.258284548042323e-06,
|
| 559 |
+
"loss": 0.2223,
|
| 560 |
+
"step": 20000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.54,
|
| 564 |
+
"eval_loss": 0.8743013739585876,
|
| 565 |
+
"eval_runtime": 2.5598,
|
| 566 |
+
"eval_samples_per_second": 1352.458,
|
| 567 |
+
"eval_steps_per_second": 7.423,
|
| 568 |
+
"step": 20000
|
| 569 |
+
}
|
| 570 |
+
],
|
| 571 |
+
"max_steps": 37238,
|
| 572 |
+
"num_train_epochs": 1,
|
| 573 |
+
"total_flos": 2.8044971311104e+17,
|
| 574 |
+
"trial_name": null,
|
| 575 |
+
"trial_params": null
|
| 576 |
+
}
|