Training complete
Browse files- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +229 -0
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6374169588088989,
|
| 4 |
+
"eval_runtime": 123.0048,
|
| 5 |
+
"eval_samples_per_second": 0.902,
|
| 6 |
+
"eval_steps_per_second": 0.228,
|
| 7 |
+
"total_flos": 4.311746565075763e+16,
|
| 8 |
+
"train_loss": 0.5089443905486001,
|
| 9 |
+
"train_runtime": 3502.9275,
|
| 10 |
+
"train_samples_per_second": 0.285,
|
| 11 |
+
"train_steps_per_second": 0.072
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6374169588088989,
|
| 4 |
+
"eval_runtime": 123.0048,
|
| 5 |
+
"eval_samples_per_second": 0.902,
|
| 6 |
+
"eval_steps_per_second": 0.228
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 4.311746565075763e+16,
|
| 4 |
+
"train_loss": 0.5089443905486001,
|
| 5 |
+
"train_runtime": 3502.9275,
|
| 6 |
+
"train_samples_per_second": 0.285,
|
| 7 |
+
"train_steps_per_second": 0.072
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,229 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 252,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.12,
|
| 13 |
+
"grad_norm": 614343.125,
|
| 14 |
+
"learning_rate": 0.0002988358809900258,
|
| 15 |
+
"loss": 1.1347,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.24,
|
| 20 |
+
"grad_norm": 96113.765625,
|
| 21 |
+
"learning_rate": 0.00029536159293436166,
|
| 22 |
+
"loss": 0.6949,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.36,
|
| 27 |
+
"grad_norm": 127274.109375,
|
| 28 |
+
"learning_rate": 0.00028963106229663063,
|
| 29 |
+
"loss": 0.6215,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.48,
|
| 34 |
+
"grad_norm": 124623.640625,
|
| 35 |
+
"learning_rate": 0.0002817332360055343,
|
| 36 |
+
"loss": 0.7171,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.6,
|
| 41 |
+
"grad_norm": 94723.8046875,
|
| 42 |
+
"learning_rate": 0.0002717907008573785,
|
| 43 |
+
"loss": 0.6622,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.71,
|
| 48 |
+
"grad_norm": 79811.34375,
|
| 49 |
+
"learning_rate": 0.0002599577807744739,
|
| 50 |
+
"loss": 0.641,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.83,
|
| 55 |
+
"grad_norm": 81335.1015625,
|
| 56 |
+
"learning_rate": 0.0002464181414529809,
|
| 57 |
+
"loss": 0.6571,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.95,
|
| 62 |
+
"grad_norm": 92003.125,
|
| 63 |
+
"learning_rate": 0.0002313819395798639,
|
| 64 |
+
"loss": 0.6412,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.0,
|
| 69 |
+
"eval_loss": 0.6449399590492249,
|
| 70 |
+
"eval_runtime": 123.5827,
|
| 71 |
+
"eval_samples_per_second": 0.898,
|
| 72 |
+
"eval_steps_per_second": 0.227,
|
| 73 |
+
"step": 84
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.07,
|
| 77 |
+
"grad_norm": 80496.6328125,
|
| 78 |
+
"learning_rate": 0.00021508256086763368,
|
| 79 |
+
"loss": 0.5549,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.19,
|
| 84 |
+
"grad_norm": 91778.5078125,
|
| 85 |
+
"learning_rate": 0.00019777299753775265,
|
| 86 |
+
"loss": 0.468,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.31,
|
| 91 |
+
"grad_norm": 91146.3828125,
|
| 92 |
+
"learning_rate": 0.0001797219214799096,
|
| 93 |
+
"loss": 0.4822,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.43,
|
| 98 |
+
"grad_norm": 103467.90625,
|
| 99 |
+
"learning_rate": 0.00016120951403796364,
|
| 100 |
+
"loss": 0.539,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.55,
|
| 105 |
+
"grad_norm": 99161.296875,
|
| 106 |
+
"learning_rate": 0.0001425231171508954,
|
| 107 |
+
"loss": 0.5263,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.67,
|
| 112 |
+
"grad_norm": 64898.0390625,
|
| 113 |
+
"learning_rate": 0.00012395277334996044,
|
| 114 |
+
"loss": 0.4564,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.79,
|
| 119 |
+
"grad_norm": 64908.39453125,
|
| 120 |
+
"learning_rate": 0.00010578672383836435,
|
| 121 |
+
"loss": 0.4603,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.9,
|
| 126 |
+
"grad_norm": 65386.96875,
|
| 127 |
+
"learning_rate": 8.830693453040829e-05,
|
| 128 |
+
"loss": 0.497,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.0,
|
| 133 |
+
"eval_loss": 0.6082237958908081,
|
| 134 |
+
"eval_runtime": 123.4189,
|
| 135 |
+
"eval_samples_per_second": 0.899,
|
| 136 |
+
"eval_steps_per_second": 0.227,
|
| 137 |
+
"step": 168
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.02,
|
| 141 |
+
"grad_norm": 60998.84765625,
|
| 142 |
+
"learning_rate": 7.17847194930753e-05,
|
| 143 |
+
"loss": 0.4428,
|
| 144 |
+
"step": 170
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.14,
|
| 148 |
+
"grad_norm": 136505.265625,
|
| 149 |
+
"learning_rate": 5.6476529721189974e-05,
|
| 150 |
+
"loss": 0.305,
|
| 151 |
+
"step": 180
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.26,
|
| 155 |
+
"grad_norm": 72723.578125,
|
| 156 |
+
"learning_rate": 4.261997261104223e-05,
|
| 157 |
+
"loss": 0.3237,
|
| 158 |
+
"step": 190
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.38,
|
| 162 |
+
"grad_norm": 89701.6875,
|
| 163 |
+
"learning_rate": 3.0430123916561672e-05,
|
| 164 |
+
"loss": 0.298,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.5,
|
| 169 |
+
"grad_norm": 72959.9453125,
|
| 170 |
+
"learning_rate": 2.009618943233419e-05,
|
| 171 |
+
"loss": 0.3401,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.62,
|
| 176 |
+
"grad_norm": 73923.734375,
|
| 177 |
+
"learning_rate": 1.1778568219438839e-05,
|
| 178 |
+
"loss": 0.3294,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.74,
|
| 183 |
+
"grad_norm": 87842.8828125,
|
| 184 |
+
"learning_rate": 5.606362957498195e-06,
|
| 185 |
+
"loss": 0.3063,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.86,
|
| 190 |
+
"grad_norm": 75173.09375,
|
| 191 |
+
"learning_rate": 1.6753760662307215e-06,
|
| 192 |
+
"loss": 0.3092,
|
| 193 |
+
"step": 240
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.98,
|
| 197 |
+
"grad_norm": 59458.79296875,
|
| 198 |
+
"learning_rate": 4.662269987756317e-08,
|
| 199 |
+
"loss": 0.3357,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.0,
|
| 204 |
+
"eval_loss": 0.6374169588088989,
|
| 205 |
+
"eval_runtime": 123.4407,
|
| 206 |
+
"eval_samples_per_second": 0.899,
|
| 207 |
+
"eval_steps_per_second": 0.227,
|
| 208 |
+
"step": 252
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.0,
|
| 212 |
+
"step": 252,
|
| 213 |
+
"total_flos": 4.311746565075763e+16,
|
| 214 |
+
"train_loss": 0.5089443905486001,
|
| 215 |
+
"train_runtime": 3502.9275,
|
| 216 |
+
"train_samples_per_second": 0.285,
|
| 217 |
+
"train_steps_per_second": 0.072
|
| 218 |
+
}
|
| 219 |
+
],
|
| 220 |
+
"logging_steps": 10,
|
| 221 |
+
"max_steps": 252,
|
| 222 |
+
"num_input_tokens_seen": 0,
|
| 223 |
+
"num_train_epochs": 3,
|
| 224 |
+
"save_steps": 10,
|
| 225 |
+
"total_flos": 4.311746565075763e+16,
|
| 226 |
+
"train_batch_size": 4,
|
| 227 |
+
"trial_name": null,
|
| 228 |
+
"trial_params": null
|
| 229 |
+
}
|