Upload 8 files
Browse files- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +247 -0
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa0ba2e0d5a838bf1a401c1cf71efe3b627dc35d67eca2c2beb40fd62983983a
|
| 3 |
+
size 33661637
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ada0ce67ce334a6d1138dbfe75e0912b0c2017290419d94e239e932b14c378f1
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:503d141bd03d55a69c94b428e328d3b060b988ace448865bb4b7d17216ac0c68
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,247 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.1660184860229492,
|
| 3 |
+
"best_model_checkpoint": "experiments/checkpoint-200",
|
| 4 |
+
"epoch": 21.867881548974943,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.73,
|
| 13 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 14 |
+
"loss": 2.5173,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.46,
|
| 19 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 20 |
+
"loss": 2.4599,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 2.19,
|
| 25 |
+
"learning_rate": 8.999999999999999e-05,
|
| 26 |
+
"loss": 2.2163,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.92,
|
| 31 |
+
"learning_rate": 0.00011999999999999999,
|
| 32 |
+
"loss": 1.7728,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 3.64,
|
| 37 |
+
"learning_rate": 0.00015,
|
| 38 |
+
"loss": 1.2772,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 3.64,
|
| 43 |
+
"eval_loss": 1.2498592138290405,
|
| 44 |
+
"eval_runtime": 8.4312,
|
| 45 |
+
"eval_samples_per_second": 23.721,
|
| 46 |
+
"eval_steps_per_second": 2.965,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.37,
|
| 51 |
+
"learning_rate": 0.00017999999999999998,
|
| 52 |
+
"loss": 1.1737,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 5.1,
|
| 57 |
+
"learning_rate": 0.00020999999999999998,
|
| 58 |
+
"loss": 1.1581,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 5.83,
|
| 63 |
+
"learning_rate": 0.00023999999999999998,
|
| 64 |
+
"loss": 1.1455,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 6.56,
|
| 69 |
+
"learning_rate": 0.00027,
|
| 70 |
+
"loss": 1.121,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 7.29,
|
| 75 |
+
"learning_rate": 0.0003,
|
| 76 |
+
"loss": 1.1123,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 7.29,
|
| 81 |
+
"eval_loss": 1.187015175819397,
|
| 82 |
+
"eval_runtime": 10.2578,
|
| 83 |
+
"eval_samples_per_second": 19.497,
|
| 84 |
+
"eval_steps_per_second": 2.437,
|
| 85 |
+
"step": 100
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 8.02,
|
| 89 |
+
"learning_rate": 0.000285,
|
| 90 |
+
"loss": 1.1124,
|
| 91 |
+
"step": 110
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 8.75,
|
| 95 |
+
"learning_rate": 0.00027,
|
| 96 |
+
"loss": 1.089,
|
| 97 |
+
"step": 120
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 9.48,
|
| 101 |
+
"learning_rate": 0.00025499999999999996,
|
| 102 |
+
"loss": 1.0991,
|
| 103 |
+
"step": 130
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 10.21,
|
| 107 |
+
"learning_rate": 0.00023999999999999998,
|
| 108 |
+
"loss": 1.0759,
|
| 109 |
+
"step": 140
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 10.93,
|
| 113 |
+
"learning_rate": 0.000225,
|
| 114 |
+
"loss": 1.0762,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 10.93,
|
| 119 |
+
"eval_loss": 1.1669864654541016,
|
| 120 |
+
"eval_runtime": 10.2044,
|
| 121 |
+
"eval_samples_per_second": 19.599,
|
| 122 |
+
"eval_steps_per_second": 2.45,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 11.66,
|
| 127 |
+
"learning_rate": 0.00020999999999999998,
|
| 128 |
+
"loss": 1.0738,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 12.39,
|
| 133 |
+
"learning_rate": 0.000195,
|
| 134 |
+
"loss": 1.0424,
|
| 135 |
+
"step": 170
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 13.12,
|
| 139 |
+
"learning_rate": 0.00017999999999999998,
|
| 140 |
+
"loss": 1.0701,
|
| 141 |
+
"step": 180
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 13.85,
|
| 145 |
+
"learning_rate": 0.000165,
|
| 146 |
+
"loss": 1.0486,
|
| 147 |
+
"step": 190
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 14.58,
|
| 151 |
+
"learning_rate": 0.00015,
|
| 152 |
+
"loss": 1.0409,
|
| 153 |
+
"step": 200
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 14.58,
|
| 157 |
+
"eval_loss": 1.1660184860229492,
|
| 158 |
+
"eval_runtime": 10.2373,
|
| 159 |
+
"eval_samples_per_second": 19.536,
|
| 160 |
+
"eval_steps_per_second": 2.442,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 15.31,
|
| 165 |
+
"learning_rate": 0.000135,
|
| 166 |
+
"loss": 1.0404,
|
| 167 |
+
"step": 210
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 16.04,
|
| 171 |
+
"learning_rate": 0.00011999999999999999,
|
| 172 |
+
"loss": 1.027,
|
| 173 |
+
"step": 220
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 16.77,
|
| 177 |
+
"learning_rate": 0.00010499999999999999,
|
| 178 |
+
"loss": 1.0251,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 17.49,
|
| 183 |
+
"learning_rate": 8.999999999999999e-05,
|
| 184 |
+
"loss": 1.0183,
|
| 185 |
+
"step": 240
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 18.22,
|
| 189 |
+
"learning_rate": 7.5e-05,
|
| 190 |
+
"loss": 1.0063,
|
| 191 |
+
"step": 250
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 18.22,
|
| 195 |
+
"eval_loss": 1.1671104431152344,
|
| 196 |
+
"eval_runtime": 10.1032,
|
| 197 |
+
"eval_samples_per_second": 19.796,
|
| 198 |
+
"eval_steps_per_second": 2.474,
|
| 199 |
+
"step": 250
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 18.95,
|
| 203 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 204 |
+
"loss": 1.0167,
|
| 205 |
+
"step": 260
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 19.68,
|
| 209 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 210 |
+
"loss": 0.9973,
|
| 211 |
+
"step": 270
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 20.41,
|
| 215 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 216 |
+
"loss": 1.0219,
|
| 217 |
+
"step": 280
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 21.14,
|
| 221 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 222 |
+
"loss": 0.993,
|
| 223 |
+
"step": 290
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 21.87,
|
| 227 |
+
"learning_rate": 0.0,
|
| 228 |
+
"loss": 0.9988,
|
| 229 |
+
"step": 300
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 21.87,
|
| 233 |
+
"eval_loss": 1.1713457107543945,
|
| 234 |
+
"eval_runtime": 10.0918,
|
| 235 |
+
"eval_samples_per_second": 19.818,
|
| 236 |
+
"eval_steps_per_second": 2.477,
|
| 237 |
+
"step": 300
|
| 238 |
+
}
|
| 239 |
+
],
|
| 240 |
+
"logging_steps": 10,
|
| 241 |
+
"max_steps": 300,
|
| 242 |
+
"num_train_epochs": 24,
|
| 243 |
+
"save_steps": 50,
|
| 244 |
+
"total_flos": 3.898336244387021e+17,
|
| 245 |
+
"trial_name": null,
|
| 246 |
+
"trial_params": null
|
| 247 |
+
}
|