End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.28307044506073,
|
| 4 |
+
"eval_runtime": 73.6303,
|
| 5 |
+
"eval_samples_per_second": 16.379,
|
| 6 |
+
"eval_steps_per_second": 2.051,
|
| 7 |
+
"total_flos": 5859923904498510.0,
|
| 8 |
+
"train_loss": 0.014361034947702911,
|
| 9 |
+
"train_runtime": 3221.5046,
|
| 10 |
+
"train_samples_per_second": 10.11,
|
| 11 |
+
"train_steps_per_second": 5.056
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.28307044506073,
|
| 4 |
+
"eval_runtime": 73.6303,
|
| 5 |
+
"eval_samples_per_second": 16.379,
|
| 6 |
+
"eval_steps_per_second": 2.051
|
| 7 |
+
}
|
runs/Aug21_16-01-46_wolf-creek/events.out.tfevents.1724281154.wolf-creek.3389603.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bd85193ee7785798006c464ddd587d9940289f7650ff992a0232d140600751d
|
| 3 |
+
size 311
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 5859923904498510.0,
|
| 4 |
+
"train_loss": 0.014361034947702911,
|
| 5 |
+
"train_runtime": 3221.5046,
|
| 6 |
+
"train_samples_per_second": 10.11,
|
| 7 |
+
"train_steps_per_second": 5.056
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,522 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.28307044506073,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-16000",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16287,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09211495946941783,
|
| 13 |
+
"grad_norm": 3.071866512298584,
|
| 14 |
+
"learning_rate": 4.84647506755097e-07,
|
| 15 |
+
"loss": 0.3448,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09211495946941783,
|
| 20 |
+
"eval_loss": 1.0838947296142578,
|
| 21 |
+
"eval_runtime": 76.5304,
|
| 22 |
+
"eval_samples_per_second": 15.772,
|
| 23 |
+
"eval_steps_per_second": 1.973,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.18422991893883567,
|
| 28 |
+
"grad_norm": 2.2681832313537598,
|
| 29 |
+
"learning_rate": 4.69295013510194e-07,
|
| 30 |
+
"loss": 0.2708,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.18422991893883567,
|
| 35 |
+
"eval_loss": 0.8948286175727844,
|
| 36 |
+
"eval_runtime": 76.7259,
|
| 37 |
+
"eval_samples_per_second": 15.731,
|
| 38 |
+
"eval_steps_per_second": 1.968,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.2763448784082535,
|
| 43 |
+
"grad_norm": 0.00790461990982294,
|
| 44 |
+
"learning_rate": 4.5394252026529107e-07,
|
| 45 |
+
"loss": 0.177,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.2763448784082535,
|
| 50 |
+
"eval_loss": 0.8396429419517517,
|
| 51 |
+
"eval_runtime": 78.5505,
|
| 52 |
+
"eval_samples_per_second": 15.366,
|
| 53 |
+
"eval_steps_per_second": 1.922,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.36845983787767134,
|
| 58 |
+
"grad_norm": 0.29154184460639954,
|
| 59 |
+
"learning_rate": 4.385900270203881e-07,
|
| 60 |
+
"loss": 0.1831,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.36845983787767134,
|
| 65 |
+
"eval_loss": 0.7720839977264404,
|
| 66 |
+
"eval_runtime": 78.1799,
|
| 67 |
+
"eval_samples_per_second": 15.439,
|
| 68 |
+
"eval_steps_per_second": 1.931,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.46057479734708917,
|
| 73 |
+
"grad_norm": 186.6424102783203,
|
| 74 |
+
"learning_rate": 4.232375337754851e-07,
|
| 75 |
+
"loss": 0.2038,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.46057479734708917,
|
| 80 |
+
"eval_loss": 0.744644284248352,
|
| 81 |
+
"eval_runtime": 76.5359,
|
| 82 |
+
"eval_samples_per_second": 15.77,
|
| 83 |
+
"eval_steps_per_second": 1.973,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.552689756816507,
|
| 88 |
+
"grad_norm": 0.009510820731520653,
|
| 89 |
+
"learning_rate": 4.0788504053058217e-07,
|
| 90 |
+
"loss": 0.1309,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.552689756816507,
|
| 95 |
+
"eval_loss": 0.7234784960746765,
|
| 96 |
+
"eval_runtime": 76.6741,
|
| 97 |
+
"eval_samples_per_second": 15.742,
|
| 98 |
+
"eval_steps_per_second": 1.969,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6448047162859248,
|
| 103 |
+
"grad_norm": 542.5567626953125,
|
| 104 |
+
"learning_rate": 3.925325472856792e-07,
|
| 105 |
+
"loss": 0.1431,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.6448047162859248,
|
| 110 |
+
"eval_loss": 0.6690872311592102,
|
| 111 |
+
"eval_runtime": 76.4873,
|
| 112 |
+
"eval_samples_per_second": 15.78,
|
| 113 |
+
"eval_steps_per_second": 1.974,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7369196757553427,
|
| 118 |
+
"grad_norm": 3.6845791339874268,
|
| 119 |
+
"learning_rate": 3.771800540407762e-07,
|
| 120 |
+
"loss": 0.1411,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7369196757553427,
|
| 125 |
+
"eval_loss": 0.643679141998291,
|
| 126 |
+
"eval_runtime": 76.4578,
|
| 127 |
+
"eval_samples_per_second": 15.786,
|
| 128 |
+
"eval_steps_per_second": 1.975,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8290346352247605,
|
| 133 |
+
"grad_norm": 0.005153482314199209,
|
| 134 |
+
"learning_rate": 3.618275607958732e-07,
|
| 135 |
+
"loss": 0.0849,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.8290346352247605,
|
| 140 |
+
"eval_loss": 0.48050031065940857,
|
| 141 |
+
"eval_runtime": 75.8694,
|
| 142 |
+
"eval_samples_per_second": 15.909,
|
| 143 |
+
"eval_steps_per_second": 1.99,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.9211495946941783,
|
| 148 |
+
"grad_norm": 2.631653070449829,
|
| 149 |
+
"learning_rate": 3.464750675509703e-07,
|
| 150 |
+
"loss": 0.1026,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.9211495946941783,
|
| 155 |
+
"eval_loss": 0.4800306260585785,
|
| 156 |
+
"eval_runtime": 75.365,
|
| 157 |
+
"eval_samples_per_second": 16.015,
|
| 158 |
+
"eval_steps_per_second": 2.004,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.013264554163596,
|
| 163 |
+
"grad_norm": 1.8651682694326155e-05,
|
| 164 |
+
"learning_rate": 3.311225743060673e-07,
|
| 165 |
+
"loss": 0.1201,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.013264554163596,
|
| 170 |
+
"eval_loss": 0.4810214042663574,
|
| 171 |
+
"eval_runtime": 74.984,
|
| 172 |
+
"eval_samples_per_second": 16.097,
|
| 173 |
+
"eval_steps_per_second": 2.014,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.105379513633014,
|
| 178 |
+
"grad_norm": 0.5627817511558533,
|
| 179 |
+
"learning_rate": 3.157700810611643e-07,
|
| 180 |
+
"loss": 0.0757,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.105379513633014,
|
| 185 |
+
"eval_loss": 0.46917983889579773,
|
| 186 |
+
"eval_runtime": 74.883,
|
| 187 |
+
"eval_samples_per_second": 16.118,
|
| 188 |
+
"eval_steps_per_second": 2.016,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.1974944731024317,
|
| 193 |
+
"grad_norm": 0.0001344973425148055,
|
| 194 |
+
"learning_rate": 3.0041758781626137e-07,
|
| 195 |
+
"loss": 0.0696,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.1974944731024317,
|
| 200 |
+
"eval_loss": 0.47636929154396057,
|
| 201 |
+
"eval_runtime": 76.4445,
|
| 202 |
+
"eval_samples_per_second": 15.789,
|
| 203 |
+
"eval_steps_per_second": 1.975,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.2896094325718497,
|
| 208 |
+
"grad_norm": 0.007305172737687826,
|
| 209 |
+
"learning_rate": 2.8506509457135833e-07,
|
| 210 |
+
"loss": 0.0911,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.2896094325718497,
|
| 215 |
+
"eval_loss": 0.4601423740386963,
|
| 216 |
+
"eval_runtime": 76.451,
|
| 217 |
+
"eval_samples_per_second": 15.788,
|
| 218 |
+
"eval_steps_per_second": 1.975,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.3817243920412676,
|
| 223 |
+
"grad_norm": 9.710428905407298e-09,
|
| 224 |
+
"learning_rate": 2.697126013264554e-07,
|
| 225 |
+
"loss": 0.0806,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.3817243920412676,
|
| 230 |
+
"eval_loss": 0.45901384949684143,
|
| 231 |
+
"eval_runtime": 74.7174,
|
| 232 |
+
"eval_samples_per_second": 16.154,
|
| 233 |
+
"eval_steps_per_second": 2.021,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4738393515106853,
|
| 238 |
+
"grad_norm": 0.008983594365417957,
|
| 239 |
+
"learning_rate": 2.5436010808155247e-07,
|
| 240 |
+
"loss": 0.088,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4738393515106853,
|
| 245 |
+
"eval_loss": 0.46537238359451294,
|
| 246 |
+
"eval_runtime": 74.5146,
|
| 247 |
+
"eval_samples_per_second": 16.198,
|
| 248 |
+
"eval_steps_per_second": 2.026,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.565954310980103,
|
| 253 |
+
"grad_norm": 1.3851042240276001e-05,
|
| 254 |
+
"learning_rate": 2.390076148366495e-07,
|
| 255 |
+
"loss": 0.0878,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.565954310980103,
|
| 260 |
+
"eval_loss": 0.4768938720226288,
|
| 261 |
+
"eval_runtime": 74.5757,
|
| 262 |
+
"eval_samples_per_second": 16.185,
|
| 263 |
+
"eval_steps_per_second": 2.025,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.658069270449521,
|
| 268 |
+
"grad_norm": 4.7310441004810855e-06,
|
| 269 |
+
"learning_rate": 2.236551215917465e-07,
|
| 270 |
+
"loss": 0.0369,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.658069270449521,
|
| 275 |
+
"eval_loss": 0.4683688282966614,
|
| 276 |
+
"eval_runtime": 74.5774,
|
| 277 |
+
"eval_samples_per_second": 16.185,
|
| 278 |
+
"eval_steps_per_second": 2.025,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.750184229918939,
|
| 283 |
+
"grad_norm": 0.005283420439809561,
|
| 284 |
+
"learning_rate": 2.083026283468435e-07,
|
| 285 |
+
"loss": 0.1034,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.750184229918939,
|
| 290 |
+
"eval_loss": 0.4715610444545746,
|
| 291 |
+
"eval_runtime": 74.6527,
|
| 292 |
+
"eval_samples_per_second": 16.168,
|
| 293 |
+
"eval_steps_per_second": 2.023,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.8422991893883567,
|
| 298 |
+
"grad_norm": 1.703993320465088,
|
| 299 |
+
"learning_rate": 1.9295013510194055e-07,
|
| 300 |
+
"loss": 0.0852,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.8422991893883567,
|
| 305 |
+
"eval_loss": 0.47203630208969116,
|
| 306 |
+
"eval_runtime": 74.6773,
|
| 307 |
+
"eval_samples_per_second": 16.163,
|
| 308 |
+
"eval_steps_per_second": 2.022,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9344141488577744,
|
| 313 |
+
"grad_norm": 8.647094205116446e-07,
|
| 314 |
+
"learning_rate": 1.7759764185703757e-07,
|
| 315 |
+
"loss": 0.0493,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.9344141488577744,
|
| 320 |
+
"eval_loss": 0.47136709094047546,
|
| 321 |
+
"eval_runtime": 74.7402,
|
| 322 |
+
"eval_samples_per_second": 16.149,
|
| 323 |
+
"eval_steps_per_second": 2.02,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.026529108327192,
|
| 328 |
+
"grad_norm": 0.0002839279477484524,
|
| 329 |
+
"learning_rate": 1.6224514861213458e-07,
|
| 330 |
+
"loss": 0.0603,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.026529108327192,
|
| 335 |
+
"eval_loss": 0.4660906493663788,
|
| 336 |
+
"eval_runtime": 74.8543,
|
| 337 |
+
"eval_samples_per_second": 16.125,
|
| 338 |
+
"eval_steps_per_second": 2.017,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.1186440677966103,
|
| 343 |
+
"grad_norm": 0.17852088809013367,
|
| 344 |
+
"learning_rate": 1.4689265536723165e-07,
|
| 345 |
+
"loss": 0.0547,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.1186440677966103,
|
| 350 |
+
"eval_loss": 0.4668720066547394,
|
| 351 |
+
"eval_runtime": 74.8609,
|
| 352 |
+
"eval_samples_per_second": 16.123,
|
| 353 |
+
"eval_steps_per_second": 2.017,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.210759027266028,
|
| 358 |
+
"grad_norm": 4.607869868777925e-06,
|
| 359 |
+
"learning_rate": 1.3154016212232866e-07,
|
| 360 |
+
"loss": 0.0793,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.210759027266028,
|
| 365 |
+
"eval_loss": 0.4664075970649719,
|
| 366 |
+
"eval_runtime": 74.7262,
|
| 367 |
+
"eval_samples_per_second": 16.152,
|
| 368 |
+
"eval_steps_per_second": 2.021,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.302449806594216,
|
| 373 |
+
"grad_norm": 0.00044986297143623233,
|
| 374 |
+
"learning_rate": 1.1625836556763061e-07,
|
| 375 |
+
"loss": 0.0415,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.302449806594216,
|
| 380 |
+
"eval_loss": 0.2888467311859131,
|
| 381 |
+
"eval_runtime": 78.2614,
|
| 382 |
+
"eval_samples_per_second": 15.41,
|
| 383 |
+
"eval_steps_per_second": 1.929,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.394547798857985,
|
| 388 |
+
"grad_norm": 4.1875861001017256e-08,
|
| 389 |
+
"learning_rate": 1.0090870019033584e-07,
|
| 390 |
+
"loss": 0.0565,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.394547798857985,
|
| 395 |
+
"eval_loss": 0.29098838567733765,
|
| 396 |
+
"eval_runtime": 78.3993,
|
| 397 |
+
"eval_samples_per_second": 15.383,
|
| 398 |
+
"eval_steps_per_second": 1.926,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.4866457911217537,
|
| 403 |
+
"grad_norm": 1.556837378302589e-05,
|
| 404 |
+
"learning_rate": 8.555903481304106e-08,
|
| 405 |
+
"loss": 0.0629,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.4866457911217537,
|
| 410 |
+
"eval_loss": 0.2888970673084259,
|
| 411 |
+
"eval_runtime": 76.7751,
|
| 412 |
+
"eval_samples_per_second": 15.708,
|
| 413 |
+
"eval_steps_per_second": 1.967,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.578743783385522,
|
| 418 |
+
"grad_norm": 0.12299621850252151,
|
| 419 |
+
"learning_rate": 7.020936943574628e-08,
|
| 420 |
+
"loss": 0.0584,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.578743783385522,
|
| 425 |
+
"eval_loss": 0.28743651509284973,
|
| 426 |
+
"eval_runtime": 76.8684,
|
| 427 |
+
"eval_samples_per_second": 15.689,
|
| 428 |
+
"eval_steps_per_second": 1.964,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.670841775649291,
|
| 433 |
+
"grad_norm": 0.25907593965530396,
|
| 434 |
+
"learning_rate": 5.485970405845152e-08,
|
| 435 |
+
"loss": 0.0582,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.670841775649291,
|
| 440 |
+
"eval_loss": 0.2863319218158722,
|
| 441 |
+
"eval_runtime": 77.0162,
|
| 442 |
+
"eval_samples_per_second": 15.659,
|
| 443 |
+
"eval_steps_per_second": 1.961,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.7629397679130596,
|
| 448 |
+
"grad_norm": 1.9590417650761083e-05,
|
| 449 |
+
"learning_rate": 3.951003868115675e-08,
|
| 450 |
+
"loss": 0.052,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.7629397679130596,
|
| 455 |
+
"eval_loss": 0.2845611274242401,
|
| 456 |
+
"eval_runtime": 76.9043,
|
| 457 |
+
"eval_samples_per_second": 15.682,
|
| 458 |
+
"eval_steps_per_second": 1.963,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.8550377601768284,
|
| 463 |
+
"grad_norm": 0.2627514600753784,
|
| 464 |
+
"learning_rate": 2.4160373303861975e-08,
|
| 465 |
+
"loss": 0.0402,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.8550377601768284,
|
| 470 |
+
"eval_loss": 0.28350818157196045,
|
| 471 |
+
"eval_runtime": 76.8887,
|
| 472 |
+
"eval_samples_per_second": 15.685,
|
| 473 |
+
"eval_steps_per_second": 1.964,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.947135752440597,
|
| 478 |
+
"grad_norm": 0.00015808168973308057,
|
| 479 |
+
"learning_rate": 8.8107079265672e-09,
|
| 480 |
+
"loss": 0.0518,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.947135752440597,
|
| 485 |
+
"eval_loss": 0.28307044506073,
|
| 486 |
+
"eval_runtime": 76.733,
|
| 487 |
+
"eval_samples_per_second": 15.717,
|
| 488 |
+
"eval_steps_per_second": 1.968,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.0,
|
| 493 |
+
"step": 16287,
|
| 494 |
+
"total_flos": 5859923904498510.0,
|
| 495 |
+
"train_loss": 0.014361034947702911,
|
| 496 |
+
"train_runtime": 3221.5046,
|
| 497 |
+
"train_samples_per_second": 10.11,
|
| 498 |
+
"train_steps_per_second": 5.056
|
| 499 |
+
}
|
| 500 |
+
],
|
| 501 |
+
"logging_steps": 500,
|
| 502 |
+
"max_steps": 16287,
|
| 503 |
+
"num_input_tokens_seen": 0,
|
| 504 |
+
"num_train_epochs": 3,
|
| 505 |
+
"save_steps": 500,
|
| 506 |
+
"stateful_callbacks": {
|
| 507 |
+
"TrainerControl": {
|
| 508 |
+
"args": {
|
| 509 |
+
"should_epoch_stop": false,
|
| 510 |
+
"should_evaluate": false,
|
| 511 |
+
"should_log": false,
|
| 512 |
+
"should_save": true,
|
| 513 |
+
"should_training_stop": true
|
| 514 |
+
},
|
| 515 |
+
"attributes": {}
|
| 516 |
+
}
|
| 517 |
+
},
|
| 518 |
+
"total_flos": 5859923904498510.0,
|
| 519 |
+
"train_batch_size": 2,
|
| 520 |
+
"trial_name": null,
|
| 521 |
+
"trial_params": null
|
| 522 |
+
}
|