Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +174 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,174 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1074,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14,
|
| 13 |
+
"learning_rate": 1.9068901303538176e-05,
|
| 14 |
+
"loss": 0.6582,
|
| 15 |
+
"step": 50
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.28,
|
| 19 |
+
"learning_rate": 1.813780260707635e-05,
|
| 20 |
+
"loss": 0.3288,
|
| 21 |
+
"step": 100
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.42,
|
| 25 |
+
"learning_rate": 1.7206703910614527e-05,
|
| 26 |
+
"loss": 0.278,
|
| 27 |
+
"step": 150
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.56,
|
| 31 |
+
"learning_rate": 1.62756052141527e-05,
|
| 32 |
+
"loss": 0.2655,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.7,
|
| 37 |
+
"learning_rate": 1.5344506517690876e-05,
|
| 38 |
+
"loss": 0.2463,
|
| 39 |
+
"step": 250
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.84,
|
| 43 |
+
"learning_rate": 1.4413407821229052e-05,
|
| 44 |
+
"loss": 0.2151,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.98,
|
| 49 |
+
"learning_rate": 1.3482309124767226e-05,
|
| 50 |
+
"loss": 0.2077,
|
| 51 |
+
"step": 350
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.12,
|
| 55 |
+
"learning_rate": 1.25512104283054e-05,
|
| 56 |
+
"loss": 0.1723,
|
| 57 |
+
"step": 400
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.26,
|
| 61 |
+
"learning_rate": 1.1620111731843577e-05,
|
| 62 |
+
"loss": 0.2021,
|
| 63 |
+
"step": 450
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.4,
|
| 67 |
+
"learning_rate": 1.0689013035381753e-05,
|
| 68 |
+
"loss": 0.1695,
|
| 69 |
+
"step": 500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.4,
|
| 73 |
+
"eval_accuracy": 0.9425142135186355,
|
| 74 |
+
"eval_loss": 0.22970134019851685,
|
| 75 |
+
"eval_runtime": 3.711,
|
| 76 |
+
"eval_samples_per_second": 426.567,
|
| 77 |
+
"eval_steps_per_second": 26.677,
|
| 78 |
+
"step": 500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.54,
|
| 82 |
+
"learning_rate": 9.757914338919926e-06,
|
| 83 |
+
"loss": 0.1799,
|
| 84 |
+
"step": 550
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.68,
|
| 88 |
+
"learning_rate": 8.826815642458101e-06,
|
| 89 |
+
"loss": 0.1664,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.82,
|
| 94 |
+
"learning_rate": 7.895716945996277e-06,
|
| 95 |
+
"loss": 0.1598,
|
| 96 |
+
"step": 650
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.96,
|
| 100 |
+
"learning_rate": 6.964618249534451e-06,
|
| 101 |
+
"loss": 0.1586,
|
| 102 |
+
"step": 700
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.09,
|
| 106 |
+
"learning_rate": 6.033519553072626e-06,
|
| 107 |
+
"loss": 0.1534,
|
| 108 |
+
"step": 750
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.23,
|
| 112 |
+
"learning_rate": 5.102420856610801e-06,
|
| 113 |
+
"loss": 0.1349,
|
| 114 |
+
"step": 800
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.37,
|
| 118 |
+
"learning_rate": 4.171322160148976e-06,
|
| 119 |
+
"loss": 0.1302,
|
| 120 |
+
"step": 850
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.51,
|
| 124 |
+
"learning_rate": 3.240223463687151e-06,
|
| 125 |
+
"loss": 0.129,
|
| 126 |
+
"step": 900
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.65,
|
| 130 |
+
"learning_rate": 2.3091247672253262e-06,
|
| 131 |
+
"loss": 0.1425,
|
| 132 |
+
"step": 950
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.79,
|
| 136 |
+
"learning_rate": 1.378026070763501e-06,
|
| 137 |
+
"loss": 0.1095,
|
| 138 |
+
"step": 1000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.79,
|
| 142 |
+
"eval_accuracy": 0.9481996209728364,
|
| 143 |
+
"eval_loss": 0.2184983193874359,
|
| 144 |
+
"eval_runtime": 3.6936,
|
| 145 |
+
"eval_samples_per_second": 428.584,
|
| 146 |
+
"eval_steps_per_second": 26.803,
|
| 147 |
+
"step": 1000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.93,
|
| 151 |
+
"learning_rate": 4.46927374301676e-07,
|
| 152 |
+
"loss": 0.1314,
|
| 153 |
+
"step": 1050
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.0,
|
| 157 |
+
"step": 1074,
|
| 158 |
+
"total_flos": 1712549613668400.0,
|
| 159 |
+
"train_loss": 0.20474217635071257,
|
| 160 |
+
"train_runtime": 279.2511,
|
| 161 |
+
"train_samples_per_second": 122.75,
|
| 162 |
+
"train_steps_per_second": 3.846
|
| 163 |
+
}
|
| 164 |
+
],
|
| 165 |
+
"logging_steps": 50,
|
| 166 |
+
"max_steps": 1074,
|
| 167 |
+
"num_input_tokens_seen": 0,
|
| 168 |
+
"num_train_epochs": 3,
|
| 169 |
+
"save_steps": 500,
|
| 170 |
+
"total_flos": 1712549613668400.0,
|
| 171 |
+
"train_batch_size": 32,
|
| 172 |
+
"trial_name": null,
|
| 173 |
+
"trial_params": null
|
| 174 |
+
}
|