Update README.md
Browse files
README.md
CHANGED
|
@@ -7,7 +7,573 @@ base_model: meta-llama/Llama-2-7b-hf
|
|
| 7 |
|
| 8 |
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
## Model Details
|
| 13 |
|
|
|
|
| 7 |
|
| 8 |
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
|
| 10 |
+
```json
|
| 11 |
+
{
|
| 12 |
+
"best_metric": 0.0680607408285141,
|
| 13 |
+
"best_model_checkpoint": "./Lora-Meta-Llama2-7b-hf-QandA_2g_v01-r2-v01\\checkpoint-19607",
|
| 14 |
+
"epoch": 38.99950273495773,
|
| 15 |
+
"eval_steps": 500,
|
| 16 |
+
"global_step": 19607,
|
| 17 |
+
"is_hyper_param_search": false,
|
| 18 |
+
"is_local_process_zero": true,
|
| 19 |
+
"is_world_process_zero": true,
|
| 20 |
+
"log_history": [
|
| 21 |
+
{
|
| 22 |
+
"epoch": 1.0,
|
| 23 |
+
"learning_rate": 9.75e-05,
|
| 24 |
+
"loss": 0.8234,
|
| 25 |
+
"step": 502
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_loss": 0.6993071436882019,
|
| 30 |
+
"eval_runtime": 196.6154,
|
| 31 |
+
"eval_samples_per_second": 2.05,
|
| 32 |
+
"eval_steps_per_second": 0.259,
|
| 33 |
+
"step": 502
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"learning_rate": 9.499501992031873e-05,
|
| 38 |
+
"loss": 0.6127,
|
| 39 |
+
"step": 1005
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"eval_loss": 0.599130392074585,
|
| 44 |
+
"eval_runtime": 196.9551,
|
| 45 |
+
"eval_samples_per_second": 2.046,
|
| 46 |
+
"eval_steps_per_second": 0.259,
|
| 47 |
+
"step": 1005
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"learning_rate": 9.249003984063745e-05,
|
| 52 |
+
"loss": 0.5345,
|
| 53 |
+
"step": 1508
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.0,
|
| 57 |
+
"eval_loss": 0.5176905989646912,
|
| 58 |
+
"eval_runtime": 198.2897,
|
| 59 |
+
"eval_samples_per_second": 2.032,
|
| 60 |
+
"eval_steps_per_second": 0.257,
|
| 61 |
+
"step": 1508
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 4.0,
|
| 65 |
+
"learning_rate": 8.998505976095618e-05,
|
| 66 |
+
"loss": 0.4679,
|
| 67 |
+
"step": 2011
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 4.0,
|
| 71 |
+
"eval_loss": 0.4569143056869507,
|
| 72 |
+
"eval_runtime": 196.7986,
|
| 73 |
+
"eval_samples_per_second": 2.048,
|
| 74 |
+
"eval_steps_per_second": 0.259,
|
| 75 |
+
"step": 2011
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.0,
|
| 79 |
+
"learning_rate": 8.748505976095617e-05,
|
| 80 |
+
"loss": 0.4086,
|
| 81 |
+
"step": 2513
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 5.0,
|
| 85 |
+
"eval_loss": 0.4007655382156372,
|
| 86 |
+
"eval_runtime": 197.0527,
|
| 87 |
+
"eval_samples_per_second": 2.045,
|
| 88 |
+
"eval_steps_per_second": 0.259,
|
| 89 |
+
"step": 2513
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"learning_rate": 8.49800796812749e-05,
|
| 94 |
+
"loss": 0.3562,
|
| 95 |
+
"step": 3016
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 6.0,
|
| 99 |
+
"eval_loss": 0.3483542799949646,
|
| 100 |
+
"eval_runtime": 198.1137,
|
| 101 |
+
"eval_samples_per_second": 2.034,
|
| 102 |
+
"eval_steps_per_second": 0.257,
|
| 103 |
+
"step": 3016
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 7.0,
|
| 107 |
+
"learning_rate": 8.247509960159363e-05,
|
| 108 |
+
"loss": 0.3143,
|
| 109 |
+
"step": 3519
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 7.0,
|
| 113 |
+
"eval_loss": 0.3145788311958313,
|
| 114 |
+
"eval_runtime": 196.7599,
|
| 115 |
+
"eval_samples_per_second": 2.048,
|
| 116 |
+
"eval_steps_per_second": 0.259,
|
| 117 |
+
"step": 3519
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 8.0,
|
| 121 |
+
"learning_rate": 7.997011952191235e-05,
|
| 122 |
+
"loss": 0.2789,
|
| 123 |
+
"step": 4022
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 8.0,
|
| 127 |
+
"eval_loss": 0.2774547338485718,
|
| 128 |
+
"eval_runtime": 195.9949,
|
| 129 |
+
"eval_samples_per_second": 2.056,
|
| 130 |
+
"eval_steps_per_second": 0.26,
|
| 131 |
+
"step": 4022
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 9.0,
|
| 135 |
+
"learning_rate": 7.747011952191235e-05,
|
| 136 |
+
"loss": 0.2506,
|
| 137 |
+
"step": 4524
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"eval_loss": 0.25074610114097595,
|
| 142 |
+
"eval_runtime": 195.9777,
|
| 143 |
+
"eval_samples_per_second": 2.056,
|
| 144 |
+
"eval_steps_per_second": 0.26,
|
| 145 |
+
"step": 4524
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 10.0,
|
| 149 |
+
"learning_rate": 7.496513944223108e-05,
|
| 150 |
+
"loss": 0.2257,
|
| 151 |
+
"step": 5027
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 10.0,
|
| 155 |
+
"eval_loss": 0.22645100951194763,
|
| 156 |
+
"eval_runtime": 195.4398,
|
| 157 |
+
"eval_samples_per_second": 2.062,
|
| 158 |
+
"eval_steps_per_second": 0.261,
|
| 159 |
+
"step": 5027
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 11.0,
|
| 163 |
+
"learning_rate": 7.24601593625498e-05,
|
| 164 |
+
"loss": 0.2031,
|
| 165 |
+
"step": 5530
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 11.0,
|
| 169 |
+
"eval_loss": 0.20663012564182281,
|
| 170 |
+
"eval_runtime": 196.0347,
|
| 171 |
+
"eval_samples_per_second": 2.056,
|
| 172 |
+
"eval_steps_per_second": 0.26,
|
| 173 |
+
"step": 5530
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 12.0,
|
| 177 |
+
"learning_rate": 6.995517928286853e-05,
|
| 178 |
+
"loss": 0.1845,
|
| 179 |
+
"step": 6033
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 12.0,
|
| 183 |
+
"eval_loss": 0.1891084611415863,
|
| 184 |
+
"eval_runtime": 195.5982,
|
| 185 |
+
"eval_samples_per_second": 2.06,
|
| 186 |
+
"eval_steps_per_second": 0.261,
|
| 187 |
+
"step": 6033
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 13.0,
|
| 191 |
+
"learning_rate": 6.745517928286854e-05,
|
| 192 |
+
"loss": 0.1691,
|
| 193 |
+
"step": 6535
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 13.0,
|
| 197 |
+
"eval_loss": 0.17209213972091675,
|
| 198 |
+
"eval_runtime": 196.259,
|
| 199 |
+
"eval_samples_per_second": 2.053,
|
| 200 |
+
"eval_steps_per_second": 0.26,
|
| 201 |
+
"step": 6535
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 14.0,
|
| 205 |
+
"learning_rate": 6.495019920318725e-05,
|
| 206 |
+
"loss": 0.1542,
|
| 207 |
+
"step": 7038
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 14.0,
|
| 211 |
+
"eval_loss": 0.15993133187294006,
|
| 212 |
+
"eval_runtime": 196.0195,
|
| 213 |
+
"eval_samples_per_second": 2.056,
|
| 214 |
+
"eval_steps_per_second": 0.26,
|
| 215 |
+
"step": 7038
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 15.0,
|
| 219 |
+
"learning_rate": 6.244521912350598e-05,
|
| 220 |
+
"loss": 0.1414,
|
| 221 |
+
"step": 7541
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 15.0,
|
| 225 |
+
"eval_loss": 0.14806699752807617,
|
| 226 |
+
"eval_runtime": 195.4909,
|
| 227 |
+
"eval_samples_per_second": 2.061,
|
| 228 |
+
"eval_steps_per_second": 0.261,
|
| 229 |
+
"step": 7541
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 16.0,
|
| 233 |
+
"learning_rate": 5.994023904382471e-05,
|
| 234 |
+
"loss": 0.1309,
|
| 235 |
+
"step": 8044
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 16.0,
|
| 239 |
+
"eval_loss": 0.13711141049861908,
|
| 240 |
+
"eval_runtime": 195.7592,
|
| 241 |
+
"eval_samples_per_second": 2.059,
|
| 242 |
+
"eval_steps_per_second": 0.261,
|
| 243 |
+
"step": 8044
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 17.0,
|
| 247 |
+
"learning_rate": 5.7440239043824705e-05,
|
| 248 |
+
"loss": 0.1222,
|
| 249 |
+
"step": 8546
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 17.0,
|
| 253 |
+
"eval_loss": 0.13089050352573395,
|
| 254 |
+
"eval_runtime": 195.6972,
|
| 255 |
+
"eval_samples_per_second": 2.059,
|
| 256 |
+
"eval_steps_per_second": 0.261,
|
| 257 |
+
"step": 8546
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 18.0,
|
| 261 |
+
"learning_rate": 5.493525896414343e-05,
|
| 262 |
+
"loss": 0.1134,
|
| 263 |
+
"step": 9049
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 18.0,
|
| 267 |
+
"eval_loss": 0.12404956668615341,
|
| 268 |
+
"eval_runtime": 195.8078,
|
| 269 |
+
"eval_samples_per_second": 2.058,
|
| 270 |
+
"eval_steps_per_second": 0.26,
|
| 271 |
+
"step": 9049
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 19.0,
|
| 275 |
+
"learning_rate": 5.243027888446216e-05,
|
| 276 |
+
"loss": 0.107,
|
| 277 |
+
"step": 9552
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 19.0,
|
| 281 |
+
"eval_loss": 0.11492674797773361,
|
| 282 |
+
"eval_runtime": 196.1937,
|
| 283 |
+
"eval_samples_per_second": 2.054,
|
| 284 |
+
"eval_steps_per_second": 0.26,
|
| 285 |
+
"step": 9552
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 20.0,
|
| 289 |
+
"learning_rate": 4.992529880478088e-05,
|
| 290 |
+
"loss": 0.1014,
|
| 291 |
+
"step": 10055
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 20.0,
|
| 295 |
+
"eval_loss": 0.1099749356508255,
|
| 296 |
+
"eval_runtime": 196.1333,
|
| 297 |
+
"eval_samples_per_second": 2.055,
|
| 298 |
+
"eval_steps_per_second": 0.26,
|
| 299 |
+
"step": 10055
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 21.0,
|
| 303 |
+
"learning_rate": 4.742529880478088e-05,
|
| 304 |
+
"loss": 0.0966,
|
| 305 |
+
"step": 10557
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 21.0,
|
| 309 |
+
"eval_loss": 0.1054670661687851,
|
| 310 |
+
"eval_runtime": 196.2688,
|
| 311 |
+
"eval_samples_per_second": 2.053,
|
| 312 |
+
"eval_steps_per_second": 0.26,
|
| 313 |
+
"step": 10557
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 22.0,
|
| 317 |
+
"learning_rate": 4.49203187250996e-05,
|
| 318 |
+
"loss": 0.0925,
|
| 319 |
+
"step": 11060
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 22.0,
|
| 323 |
+
"eval_loss": 0.1019241139292717,
|
| 324 |
+
"eval_runtime": 195.6941,
|
| 325 |
+
"eval_samples_per_second": 2.059,
|
| 326 |
+
"eval_steps_per_second": 0.261,
|
| 327 |
+
"step": 11060
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 23.0,
|
| 331 |
+
"learning_rate": 4.241533864541833e-05,
|
| 332 |
+
"loss": 0.0883,
|
| 333 |
+
"step": 11563
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 23.0,
|
| 337 |
+
"eval_loss": 0.09764768928289413,
|
| 338 |
+
"eval_runtime": 196.6317,
|
| 339 |
+
"eval_samples_per_second": 2.05,
|
| 340 |
+
"eval_steps_per_second": 0.259,
|
| 341 |
+
"step": 11563
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 24.0,
|
| 345 |
+
"learning_rate": 3.9910358565737054e-05,
|
| 346 |
+
"loss": 0.085,
|
| 347 |
+
"step": 12066
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 24.0,
|
| 351 |
+
"eval_loss": 0.09518074989318848,
|
| 352 |
+
"eval_runtime": 196.0702,
|
| 353 |
+
"eval_samples_per_second": 2.055,
|
| 354 |
+
"eval_steps_per_second": 0.26,
|
| 355 |
+
"step": 12066
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 25.0,
|
| 359 |
+
"learning_rate": 3.7410358565737055e-05,
|
| 360 |
+
"loss": 0.0832,
|
| 361 |
+
"step": 12568
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 25.0,
|
| 365 |
+
"eval_loss": 0.09052950888872147,
|
| 366 |
+
"eval_runtime": 196.4373,
|
| 367 |
+
"eval_samples_per_second": 2.052,
|
| 368 |
+
"eval_steps_per_second": 0.26,
|
| 369 |
+
"step": 12568
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 26.0,
|
| 373 |
+
"learning_rate": 3.490537848605578e-05,
|
| 374 |
+
"loss": 0.0795,
|
| 375 |
+
"step": 13071
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 26.0,
|
| 379 |
+
"eval_loss": 0.0874876081943512,
|
| 380 |
+
"eval_runtime": 196.0257,
|
| 381 |
+
"eval_samples_per_second": 2.056,
|
| 382 |
+
"eval_steps_per_second": 0.26,
|
| 383 |
+
"step": 13071
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 27.0,
|
| 387 |
+
"learning_rate": 3.24003984063745e-05,
|
| 388 |
+
"loss": 0.078,
|
| 389 |
+
"step": 13574
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 27.0,
|
| 393 |
+
"eval_loss": 0.08521231263875961,
|
| 394 |
+
"eval_runtime": 195.7721,
|
| 395 |
+
"eval_samples_per_second": 2.059,
|
| 396 |
+
"eval_steps_per_second": 0.261,
|
| 397 |
+
"step": 13574
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 28.0,
|
| 401 |
+
"learning_rate": 2.989541832669323e-05,
|
| 402 |
+
"loss": 0.0761,
|
| 403 |
+
"step": 14077
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 28.0,
|
| 407 |
+
"eval_loss": 0.08261983096599579,
|
| 408 |
+
"eval_runtime": 195.9135,
|
| 409 |
+
"eval_samples_per_second": 2.057,
|
| 410 |
+
"eval_steps_per_second": 0.26,
|
| 411 |
+
"step": 14077
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 29.0,
|
| 415 |
+
"learning_rate": 2.739541832669323e-05,
|
| 416 |
+
"loss": 0.0748,
|
| 417 |
+
"step": 14579
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 29.0,
|
| 421 |
+
"eval_loss": 0.08173193037509918,
|
| 422 |
+
"eval_runtime": 196.0573,
|
| 423 |
+
"eval_samples_per_second": 2.056,
|
| 424 |
+
"eval_steps_per_second": 0.26,
|
| 425 |
+
"step": 14579
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 30.0,
|
| 429 |
+
"learning_rate": 2.4890438247011953e-05,
|
| 430 |
+
"loss": 0.0727,
|
| 431 |
+
"step": 15082
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 30.0,
|
| 435 |
+
"eval_loss": 0.07807794213294983,
|
| 436 |
+
"eval_runtime": 196.0191,
|
| 437 |
+
"eval_samples_per_second": 2.056,
|
| 438 |
+
"eval_steps_per_second": 0.26,
|
| 439 |
+
"step": 15082
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 31.0,
|
| 443 |
+
"learning_rate": 2.2385458167330677e-05,
|
| 444 |
+
"loss": 0.0712,
|
| 445 |
+
"step": 15585
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 31.0,
|
| 449 |
+
"eval_loss": 0.07727421820163727,
|
| 450 |
+
"eval_runtime": 196.043,
|
| 451 |
+
"eval_samples_per_second": 2.056,
|
| 452 |
+
"eval_steps_per_second": 0.26,
|
| 453 |
+
"step": 15585
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 32.0,
|
| 457 |
+
"learning_rate": 1.9880478087649404e-05,
|
| 458 |
+
"loss": 0.07,
|
| 459 |
+
"step": 16088
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 32.0,
|
| 463 |
+
"eval_loss": 0.07521162927150726,
|
| 464 |
+
"eval_runtime": 196.5081,
|
| 465 |
+
"eval_samples_per_second": 2.051,
|
| 466 |
+
"eval_steps_per_second": 0.26,
|
| 467 |
+
"step": 16088
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 33.0,
|
| 471 |
+
"learning_rate": 1.7380478087649405e-05,
|
| 472 |
+
"loss": 0.069,
|
| 473 |
+
"step": 16590
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 33.0,
|
| 477 |
+
"eval_loss": 0.07360897213220596,
|
| 478 |
+
"eval_runtime": 196.0187,
|
| 479 |
+
"eval_samples_per_second": 2.056,
|
| 480 |
+
"eval_steps_per_second": 0.26,
|
| 481 |
+
"step": 16590
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 34.0,
|
| 485 |
+
"learning_rate": 1.4875498007968128e-05,
|
| 486 |
+
"loss": 0.0682,
|
| 487 |
+
"step": 17093
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 34.0,
|
| 491 |
+
"eval_loss": 0.07235261797904968,
|
| 492 |
+
"eval_runtime": 196.2571,
|
| 493 |
+
"eval_samples_per_second": 2.053,
|
| 494 |
+
"eval_steps_per_second": 0.26,
|
| 495 |
+
"step": 17093
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 35.0,
|
| 499 |
+
"learning_rate": 1.2370517928286854e-05,
|
| 500 |
+
"loss": 0.0671,
|
| 501 |
+
"step": 17596
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 35.0,
|
| 505 |
+
"eval_loss": 0.07117172330617905,
|
| 506 |
+
"eval_runtime": 197.0583,
|
| 507 |
+
"eval_samples_per_second": 2.045,
|
| 508 |
+
"eval_steps_per_second": 0.259,
|
| 509 |
+
"step": 17596
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 36.0,
|
| 513 |
+
"learning_rate": 9.86553784860558e-06,
|
| 514 |
+
"loss": 0.0663,
|
| 515 |
+
"step": 18099
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 36.0,
|
| 519 |
+
"eval_loss": 0.07005689293146133,
|
| 520 |
+
"eval_runtime": 196.2214,
|
| 521 |
+
"eval_samples_per_second": 2.054,
|
| 522 |
+
"eval_steps_per_second": 0.26,
|
| 523 |
+
"step": 18099
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 37.0,
|
| 527 |
+
"learning_rate": 7.365537848605578e-06,
|
| 528 |
+
"loss": 0.0656,
|
| 529 |
+
"step": 18601
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 37.0,
|
| 533 |
+
"eval_loss": 0.06923193484544754,
|
| 534 |
+
"eval_runtime": 195.9767,
|
| 535 |
+
"eval_samples_per_second": 2.056,
|
| 536 |
+
"eval_steps_per_second": 0.26,
|
| 537 |
+
"step": 18601
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 38.0,
|
| 541 |
+
"learning_rate": 4.860557768924303e-06,
|
| 542 |
+
"loss": 0.0648,
|
| 543 |
+
"step": 19104
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 38.0,
|
| 547 |
+
"eval_loss": 0.06851127743721008,
|
| 548 |
+
"eval_runtime": 196.8249,
|
| 549 |
+
"eval_samples_per_second": 2.048,
|
| 550 |
+
"eval_steps_per_second": 0.259,
|
| 551 |
+
"step": 19104
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 39.0,
|
| 555 |
+
"learning_rate": 2.355577689243028e-06,
|
| 556 |
+
"loss": 0.064,
|
| 557 |
+
"step": 19607
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 39.0,
|
| 561 |
+
"eval_loss": 0.0680607408285141,
|
| 562 |
+
"eval_runtime": 195.9876,
|
| 563 |
+
"eval_samples_per_second": 2.056,
|
| 564 |
+
"eval_steps_per_second": 0.26,
|
| 565 |
+
"step": 19607
|
| 566 |
+
}
|
| 567 |
+
],
|
| 568 |
+
"logging_steps": 500,
|
| 569 |
+
"max_steps": 20080,
|
| 570 |
+
"num_train_epochs": 40,
|
| 571 |
+
"save_steps": 500,
|
| 572 |
+
"total_flos": 4.6336747921934746e+17,
|
| 573 |
+
"trial_name": null,
|
| 574 |
+
"trial_params": null
|
| 575 |
+
}
|
| 576 |
+
```
|
| 577 |
|
| 578 |
## Model Details
|
| 579 |
|