End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +483 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.24,
|
| 3 |
+
"eval_accuracy": 0.9583333333333334,
|
| 4 |
+
"eval_loss": 0.15978051722049713,
|
| 5 |
+
"eval_runtime": 59.6275,
|
| 6 |
+
"eval_samples_per_second": 1.61,
|
| 7 |
+
"eval_steps_per_second": 0.268
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.24,
|
| 3 |
+
"eval_accuracy": 0.9583333333333334,
|
| 4 |
+
"eval_loss": 0.15978051722049713,
|
| 5 |
+
"eval_runtime": 59.6275,
|
| 6 |
+
"eval_samples_per_second": 1.61,
|
| 7 |
+
"eval_steps_per_second": 0.268
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,483 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9473684210526315,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-my-dataset-4-epochs\\checkpoint-572",
|
| 4 |
+
"epoch": 3.2447552447552446,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 572,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 6.4546613693237305,
|
| 14 |
+
"learning_rate": 8.620689655172414e-06,
|
| 15 |
+
"loss": 1.661,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 5.571170330047607,
|
| 21 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 22 |
+
"loss": 1.6084,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05,
|
| 27 |
+
"grad_norm": 6.379495143890381,
|
| 28 |
+
"learning_rate": 2.5862068965517244e-05,
|
| 29 |
+
"loss": 1.5554,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07,
|
| 34 |
+
"grad_norm": 5.455715656280518,
|
| 35 |
+
"learning_rate": 3.4482758620689657e-05,
|
| 36 |
+
"loss": 1.4752,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09,
|
| 41 |
+
"grad_norm": 8.574926376342773,
|
| 42 |
+
"learning_rate": 4.3103448275862066e-05,
|
| 43 |
+
"loss": 1.4367,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 7.630256652832031,
|
| 49 |
+
"learning_rate": 4.980544747081712e-05,
|
| 50 |
+
"loss": 1.2503,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12,
|
| 55 |
+
"grad_norm": 13.884824752807617,
|
| 56 |
+
"learning_rate": 4.883268482490273e-05,
|
| 57 |
+
"loss": 1.1133,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14,
|
| 62 |
+
"grad_norm": 11.933884620666504,
|
| 63 |
+
"learning_rate": 4.7859922178988334e-05,
|
| 64 |
+
"loss": 1.0921,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"grad_norm": 11.359259605407715,
|
| 70 |
+
"learning_rate": 4.688715953307393e-05,
|
| 71 |
+
"loss": 0.7924,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.17,
|
| 76 |
+
"grad_norm": 12.164495468139648,
|
| 77 |
+
"learning_rate": 4.591439688715953e-05,
|
| 78 |
+
"loss": 0.8217,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.19,
|
| 83 |
+
"grad_norm": 4.461577415466309,
|
| 84 |
+
"learning_rate": 4.494163424124514e-05,
|
| 85 |
+
"loss": 0.4345,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.21,
|
| 90 |
+
"grad_norm": 14.569128036499023,
|
| 91 |
+
"learning_rate": 4.396887159533074e-05,
|
| 92 |
+
"loss": 0.5056,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.23,
|
| 97 |
+
"grad_norm": 12.43501091003418,
|
| 98 |
+
"learning_rate": 4.299610894941635e-05,
|
| 99 |
+
"loss": 0.4189,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.24,
|
| 104 |
+
"grad_norm": 17.019834518432617,
|
| 105 |
+
"learning_rate": 4.202334630350195e-05,
|
| 106 |
+
"loss": 0.4072,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.25,
|
| 111 |
+
"eval_accuracy": 0.7473684210526316,
|
| 112 |
+
"eval_loss": 0.6893969178199768,
|
| 113 |
+
"eval_runtime": 117.3206,
|
| 114 |
+
"eval_samples_per_second": 1.619,
|
| 115 |
+
"eval_steps_per_second": 0.273,
|
| 116 |
+
"step": 144
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.01,
|
| 120 |
+
"grad_norm": 4.544285774230957,
|
| 121 |
+
"learning_rate": 4.105058365758755e-05,
|
| 122 |
+
"loss": 0.425,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.03,
|
| 127 |
+
"grad_norm": 4.996816635131836,
|
| 128 |
+
"learning_rate": 4.007782101167315e-05,
|
| 129 |
+
"loss": 0.2163,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.05,
|
| 134 |
+
"grad_norm": 8.629969596862793,
|
| 135 |
+
"learning_rate": 3.910505836575876e-05,
|
| 136 |
+
"loss": 0.3379,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.06,
|
| 141 |
+
"grad_norm": 11.086934089660645,
|
| 142 |
+
"learning_rate": 3.813229571984436e-05,
|
| 143 |
+
"loss": 0.1983,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.08,
|
| 148 |
+
"grad_norm": 17.69314193725586,
|
| 149 |
+
"learning_rate": 3.715953307392996e-05,
|
| 150 |
+
"loss": 0.3418,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.1,
|
| 155 |
+
"grad_norm": 3.8406598567962646,
|
| 156 |
+
"learning_rate": 3.618677042801556e-05,
|
| 157 |
+
"loss": 0.1979,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.12,
|
| 162 |
+
"grad_norm": 8.322322845458984,
|
| 163 |
+
"learning_rate": 3.5214007782101175e-05,
|
| 164 |
+
"loss": 0.522,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.13,
|
| 169 |
+
"grad_norm": 26.0389461517334,
|
| 170 |
+
"learning_rate": 3.4241245136186774e-05,
|
| 171 |
+
"loss": 0.2249,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.15,
|
| 176 |
+
"grad_norm": 14.741497039794922,
|
| 177 |
+
"learning_rate": 3.326848249027237e-05,
|
| 178 |
+
"loss": 0.2022,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.17,
|
| 183 |
+
"grad_norm": 25.44049072265625,
|
| 184 |
+
"learning_rate": 3.229571984435798e-05,
|
| 185 |
+
"loss": 0.3132,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.19,
|
| 190 |
+
"grad_norm": 6.259922981262207,
|
| 191 |
+
"learning_rate": 3.1322957198443584e-05,
|
| 192 |
+
"loss": 0.308,
|
| 193 |
+
"step": 250
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.2,
|
| 197 |
+
"grad_norm": 3.718748092651367,
|
| 198 |
+
"learning_rate": 3.0350194552529183e-05,
|
| 199 |
+
"loss": 0.1048,
|
| 200 |
+
"step": 260
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.22,
|
| 204 |
+
"grad_norm": 9.242581367492676,
|
| 205 |
+
"learning_rate": 2.937743190661479e-05,
|
| 206 |
+
"loss": 0.1878,
|
| 207 |
+
"step": 270
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.24,
|
| 211 |
+
"grad_norm": 0.549322783946991,
|
| 212 |
+
"learning_rate": 2.8404669260700388e-05,
|
| 213 |
+
"loss": 0.1778,
|
| 214 |
+
"step": 280
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.25,
|
| 218 |
+
"eval_accuracy": 0.9105263157894737,
|
| 219 |
+
"eval_loss": 0.2762860655784607,
|
| 220 |
+
"eval_runtime": 120.6902,
|
| 221 |
+
"eval_samples_per_second": 1.574,
|
| 222 |
+
"eval_steps_per_second": 0.265,
|
| 223 |
+
"step": 288
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.0,
|
| 227 |
+
"grad_norm": 0.7821664810180664,
|
| 228 |
+
"learning_rate": 2.7431906614785997e-05,
|
| 229 |
+
"loss": 0.2725,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.02,
|
| 234 |
+
"grad_norm": 3.510664939880371,
|
| 235 |
+
"learning_rate": 2.6459143968871596e-05,
|
| 236 |
+
"loss": 0.0888,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.04,
|
| 241 |
+
"grad_norm": 0.5098416805267334,
|
| 242 |
+
"learning_rate": 2.54863813229572e-05,
|
| 243 |
+
"loss": 0.0764,
|
| 244 |
+
"step": 310
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.06,
|
| 248 |
+
"grad_norm": 1.5534539222717285,
|
| 249 |
+
"learning_rate": 2.45136186770428e-05,
|
| 250 |
+
"loss": 0.0346,
|
| 251 |
+
"step": 320
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.07,
|
| 255 |
+
"grad_norm": 6.638437271118164,
|
| 256 |
+
"learning_rate": 2.3540856031128406e-05,
|
| 257 |
+
"loss": 0.0397,
|
| 258 |
+
"step": 330
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.09,
|
| 262 |
+
"grad_norm": 35.592262268066406,
|
| 263 |
+
"learning_rate": 2.256809338521401e-05,
|
| 264 |
+
"loss": 0.2642,
|
| 265 |
+
"step": 340
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.11,
|
| 269 |
+
"grad_norm": 0.5243805050849915,
|
| 270 |
+
"learning_rate": 2.1595330739299615e-05,
|
| 271 |
+
"loss": 0.1475,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.13,
|
| 276 |
+
"grad_norm": 16.06882095336914,
|
| 277 |
+
"learning_rate": 2.0622568093385214e-05,
|
| 278 |
+
"loss": 0.0525,
|
| 279 |
+
"step": 360
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.14,
|
| 283 |
+
"grad_norm": 21.048095703125,
|
| 284 |
+
"learning_rate": 1.964980544747082e-05,
|
| 285 |
+
"loss": 0.0923,
|
| 286 |
+
"step": 370
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.16,
|
| 290 |
+
"grad_norm": 0.9029656648635864,
|
| 291 |
+
"learning_rate": 1.867704280155642e-05,
|
| 292 |
+
"loss": 0.1186,
|
| 293 |
+
"step": 380
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.18,
|
| 297 |
+
"grad_norm": 19.88978385925293,
|
| 298 |
+
"learning_rate": 1.7704280155642024e-05,
|
| 299 |
+
"loss": 0.068,
|
| 300 |
+
"step": 390
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.2,
|
| 304 |
+
"grad_norm": 1.3962284326553345,
|
| 305 |
+
"learning_rate": 1.6731517509727626e-05,
|
| 306 |
+
"loss": 0.0854,
|
| 307 |
+
"step": 400
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.21,
|
| 311 |
+
"grad_norm": 8.434904098510742,
|
| 312 |
+
"learning_rate": 1.575875486381323e-05,
|
| 313 |
+
"loss": 0.0269,
|
| 314 |
+
"step": 410
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.23,
|
| 318 |
+
"grad_norm": 0.28366413712501526,
|
| 319 |
+
"learning_rate": 1.4785992217898833e-05,
|
| 320 |
+
"loss": 0.1492,
|
| 321 |
+
"step": 420
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.25,
|
| 325 |
+
"grad_norm": 9.057991027832031,
|
| 326 |
+
"learning_rate": 1.3813229571984437e-05,
|
| 327 |
+
"loss": 0.1602,
|
| 328 |
+
"step": 430
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.25,
|
| 332 |
+
"eval_accuracy": 0.8947368421052632,
|
| 333 |
+
"eval_loss": 0.2830742597579956,
|
| 334 |
+
"eval_runtime": 127.7774,
|
| 335 |
+
"eval_samples_per_second": 1.487,
|
| 336 |
+
"eval_steps_per_second": 0.25,
|
| 337 |
+
"step": 432
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.01,
|
| 341 |
+
"grad_norm": 0.4281996190547943,
|
| 342 |
+
"learning_rate": 1.2840466926070038e-05,
|
| 343 |
+
"loss": 0.0725,
|
| 344 |
+
"step": 440
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.03,
|
| 348 |
+
"grad_norm": 0.3639540374279022,
|
| 349 |
+
"learning_rate": 1.1867704280155642e-05,
|
| 350 |
+
"loss": 0.0253,
|
| 351 |
+
"step": 450
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.05,
|
| 355 |
+
"grad_norm": 0.6328150629997253,
|
| 356 |
+
"learning_rate": 1.0894941634241246e-05,
|
| 357 |
+
"loss": 0.0083,
|
| 358 |
+
"step": 460
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.07,
|
| 362 |
+
"grad_norm": 0.0523068867623806,
|
| 363 |
+
"learning_rate": 9.922178988326848e-06,
|
| 364 |
+
"loss": 0.0335,
|
| 365 |
+
"step": 470
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.08,
|
| 369 |
+
"grad_norm": 0.029841965064406395,
|
| 370 |
+
"learning_rate": 8.949416342412452e-06,
|
| 371 |
+
"loss": 0.0033,
|
| 372 |
+
"step": 480
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.1,
|
| 376 |
+
"grad_norm": 0.06523406505584717,
|
| 377 |
+
"learning_rate": 7.976653696498054e-06,
|
| 378 |
+
"loss": 0.0622,
|
| 379 |
+
"step": 490
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 3.12,
|
| 383 |
+
"grad_norm": 6.156842231750488,
|
| 384 |
+
"learning_rate": 7.003891050583658e-06,
|
| 385 |
+
"loss": 0.0621,
|
| 386 |
+
"step": 500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.14,
|
| 390 |
+
"grad_norm": 0.4030187129974365,
|
| 391 |
+
"learning_rate": 6.031128404669261e-06,
|
| 392 |
+
"loss": 0.0478,
|
| 393 |
+
"step": 510
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.15,
|
| 397 |
+
"grad_norm": 7.486963748931885,
|
| 398 |
+
"learning_rate": 5.058365758754864e-06,
|
| 399 |
+
"loss": 0.0057,
|
| 400 |
+
"step": 520
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.17,
|
| 404 |
+
"grad_norm": 21.099483489990234,
|
| 405 |
+
"learning_rate": 4.085603112840467e-06,
|
| 406 |
+
"loss": 0.0575,
|
| 407 |
+
"step": 530
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 3.19,
|
| 411 |
+
"grad_norm": 0.5362010598182678,
|
| 412 |
+
"learning_rate": 3.11284046692607e-06,
|
| 413 |
+
"loss": 0.1306,
|
| 414 |
+
"step": 540
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 3.21,
|
| 418 |
+
"grad_norm": 2.0333070755004883,
|
| 419 |
+
"learning_rate": 2.1400778210116734e-06,
|
| 420 |
+
"loss": 0.0212,
|
| 421 |
+
"step": 550
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.22,
|
| 425 |
+
"grad_norm": 0.11834029853343964,
|
| 426 |
+
"learning_rate": 1.1673151750972764e-06,
|
| 427 |
+
"loss": 0.0824,
|
| 428 |
+
"step": 560
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 3.24,
|
| 432 |
+
"grad_norm": 0.06718810647726059,
|
| 433 |
+
"learning_rate": 1.9455252918287938e-07,
|
| 434 |
+
"loss": 0.0134,
|
| 435 |
+
"step": 570
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 3.24,
|
| 439 |
+
"eval_accuracy": 0.9473684210526315,
|
| 440 |
+
"eval_loss": 0.18670256435871124,
|
| 441 |
+
"eval_runtime": 127.8371,
|
| 442 |
+
"eval_samples_per_second": 1.486,
|
| 443 |
+
"eval_steps_per_second": 0.25,
|
| 444 |
+
"step": 572
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 3.24,
|
| 448 |
+
"step": 572,
|
| 449 |
+
"total_flos": 4.2579166172730163e+18,
|
| 450 |
+
"train_loss": 0.3607311663168997,
|
| 451 |
+
"train_runtime": 3239.8454,
|
| 452 |
+
"train_samples_per_second": 1.059,
|
| 453 |
+
"train_steps_per_second": 0.177
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.24,
|
| 457 |
+
"eval_accuracy": 0.9583333333333334,
|
| 458 |
+
"eval_loss": 0.15978054702281952,
|
| 459 |
+
"eval_runtime": 64.9438,
|
| 460 |
+
"eval_samples_per_second": 1.478,
|
| 461 |
+
"eval_steps_per_second": 0.246,
|
| 462 |
+
"step": 572
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 3.24,
|
| 466 |
+
"eval_accuracy": 0.9583333333333334,
|
| 467 |
+
"eval_loss": 0.15978051722049713,
|
| 468 |
+
"eval_runtime": 59.6275,
|
| 469 |
+
"eval_samples_per_second": 1.61,
|
| 470 |
+
"eval_steps_per_second": 0.268,
|
| 471 |
+
"step": 572
|
| 472 |
+
}
|
| 473 |
+
],
|
| 474 |
+
"logging_steps": 10,
|
| 475 |
+
"max_steps": 572,
|
| 476 |
+
"num_input_tokens_seen": 0,
|
| 477 |
+
"num_train_epochs": 9223372036854775807,
|
| 478 |
+
"save_steps": 500,
|
| 479 |
+
"total_flos": 4.2579166172730163e+18,
|
| 480 |
+
"train_batch_size": 6,
|
| 481 |
+
"trial_name": null,
|
| 482 |
+
"trial_params": null
|
| 483 |
+
}
|