Commit
·
bedf785
1
Parent(s):
f264eaf
Upload trainer_state.json
Browse files- trainer_state.json +206 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,206 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.860524091293322,
|
| 5 |
+
"global_step": 11500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.21,
|
| 12 |
+
"learning_rate": 2.8732037193575658e-05,
|
| 13 |
+
"loss": 3.5507,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.42,
|
| 18 |
+
"learning_rate": 2.7464074387151312e-05,
|
| 19 |
+
"loss": 3.139,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.63,
|
| 24 |
+
"learning_rate": 2.6196111580726966e-05,
|
| 25 |
+
"loss": 3.0681,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.85,
|
| 30 |
+
"learning_rate": 2.492814877430262e-05,
|
| 31 |
+
"loss": 3.0193,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"eval_gen_len": 7.8559,
|
| 37 |
+
"eval_loss": 2.918958902359009,
|
| 38 |
+
"eval_rouge1": 48.6486,
|
| 39 |
+
"eval_rouge2": 0.0,
|
| 40 |
+
"eval_rougeL": 48.6486,
|
| 41 |
+
"eval_rougeLsum": 48.6486,
|
| 42 |
+
"eval_runtime": 1.4313,
|
| 43 |
+
"eval_samples_per_second": 77.553,
|
| 44 |
+
"eval_steps_per_second": 4.891,
|
| 45 |
+
"step": 2366
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.06,
|
| 49 |
+
"learning_rate": 2.3660185967878277e-05,
|
| 50 |
+
"loss": 2.9837,
|
| 51 |
+
"step": 2500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.27,
|
| 55 |
+
"learning_rate": 2.239222316145393e-05,
|
| 56 |
+
"loss": 2.9458,
|
| 57 |
+
"step": 3000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.48,
|
| 61 |
+
"learning_rate": 2.112426035502959e-05,
|
| 62 |
+
"loss": 2.9247,
|
| 63 |
+
"step": 3500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.69,
|
| 67 |
+
"learning_rate": 1.985629754860524e-05,
|
| 68 |
+
"loss": 2.9004,
|
| 69 |
+
"step": 4000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.9,
|
| 73 |
+
"learning_rate": 1.8588334742180896e-05,
|
| 74 |
+
"loss": 2.9038,
|
| 75 |
+
"step": 4500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 2.0,
|
| 79 |
+
"eval_gen_len": 7.7387,
|
| 80 |
+
"eval_loss": 2.838879346847534,
|
| 81 |
+
"eval_rouge1": 48.6486,
|
| 82 |
+
"eval_rouge2": 0.0,
|
| 83 |
+
"eval_rougeL": 48.6486,
|
| 84 |
+
"eval_rougeLsum": 48.6486,
|
| 85 |
+
"eval_runtime": 1.069,
|
| 86 |
+
"eval_samples_per_second": 103.833,
|
| 87 |
+
"eval_steps_per_second": 6.548,
|
| 88 |
+
"step": 4732
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.11,
|
| 92 |
+
"learning_rate": 1.732037193575655e-05,
|
| 93 |
+
"loss": 2.8876,
|
| 94 |
+
"step": 5000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.32,
|
| 98 |
+
"learning_rate": 1.6052409129332207e-05,
|
| 99 |
+
"loss": 2.8646,
|
| 100 |
+
"step": 5500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.54,
|
| 104 |
+
"learning_rate": 1.4784446322907861e-05,
|
| 105 |
+
"loss": 2.8488,
|
| 106 |
+
"step": 6000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.75,
|
| 110 |
+
"learning_rate": 1.3516483516483517e-05,
|
| 111 |
+
"loss": 2.8431,
|
| 112 |
+
"step": 6500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.96,
|
| 116 |
+
"learning_rate": 1.224852071005917e-05,
|
| 117 |
+
"loss": 2.8369,
|
| 118 |
+
"step": 7000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 3.0,
|
| 122 |
+
"eval_gen_len": 7.8468,
|
| 123 |
+
"eval_loss": 2.808067560195923,
|
| 124 |
+
"eval_rouge1": 48.6486,
|
| 125 |
+
"eval_rouge2": 0.0,
|
| 126 |
+
"eval_rougeL": 48.6486,
|
| 127 |
+
"eval_rougeLsum": 48.6486,
|
| 128 |
+
"eval_runtime": 1.0729,
|
| 129 |
+
"eval_samples_per_second": 103.457,
|
| 130 |
+
"eval_steps_per_second": 6.524,
|
| 131 |
+
"step": 7098
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.17,
|
| 135 |
+
"learning_rate": 1.0980557903634826e-05,
|
| 136 |
+
"loss": 2.8347,
|
| 137 |
+
"step": 7500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 3.38,
|
| 141 |
+
"learning_rate": 9.712595097210482e-06,
|
| 142 |
+
"loss": 2.82,
|
| 143 |
+
"step": 8000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.59,
|
| 147 |
+
"learning_rate": 8.444632290786136e-06,
|
| 148 |
+
"loss": 2.826,
|
| 149 |
+
"step": 8500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.8,
|
| 153 |
+
"learning_rate": 7.1766694843617924e-06,
|
| 154 |
+
"loss": 2.7988,
|
| 155 |
+
"step": 9000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 4.0,
|
| 159 |
+
"eval_gen_len": 8.027,
|
| 160 |
+
"eval_loss": 2.785550832748413,
|
| 161 |
+
"eval_rouge1": 48.6486,
|
| 162 |
+
"eval_rouge2": 0.0,
|
| 163 |
+
"eval_rougeL": 48.6486,
|
| 164 |
+
"eval_rougeLsum": 48.6486,
|
| 165 |
+
"eval_runtime": 1.0854,
|
| 166 |
+
"eval_samples_per_second": 102.268,
|
| 167 |
+
"eval_steps_per_second": 6.449,
|
| 168 |
+
"step": 9464
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 4.02,
|
| 172 |
+
"learning_rate": 5.908706677937447e-06,
|
| 173 |
+
"loss": 2.8132,
|
| 174 |
+
"step": 9500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.23,
|
| 178 |
+
"learning_rate": 4.640743871513102e-06,
|
| 179 |
+
"loss": 2.7946,
|
| 180 |
+
"step": 10000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.44,
|
| 184 |
+
"learning_rate": 3.3727810650887576e-06,
|
| 185 |
+
"loss": 2.8009,
|
| 186 |
+
"step": 10500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 4.65,
|
| 190 |
+
"learning_rate": 2.1048182586644128e-06,
|
| 191 |
+
"loss": 2.7986,
|
| 192 |
+
"step": 11000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.86,
|
| 196 |
+
"learning_rate": 8.368554522400676e-07,
|
| 197 |
+
"loss": 2.7982,
|
| 198 |
+
"step": 11500
|
| 199 |
+
}
|
| 200 |
+
],
|
| 201 |
+
"max_steps": 11830,
|
| 202 |
+
"num_train_epochs": 5,
|
| 203 |
+
"total_flos": 5833380202217472.0,
|
| 204 |
+
"trial_name": null,
|
| 205 |
+
"trial_params": null
|
| 206 |
+
}
|