Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +186 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1176,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.13,
|
| 13 |
+
"learning_rate": 1.914965986394558e-05,
|
| 14 |
+
"loss": 0.3064,
|
| 15 |
+
"step": 50
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.26,
|
| 19 |
+
"learning_rate": 1.8299319727891158e-05,
|
| 20 |
+
"loss": 0.0263,
|
| 21 |
+
"step": 100
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.38,
|
| 25 |
+
"learning_rate": 1.7448979591836738e-05,
|
| 26 |
+
"loss": 0.0175,
|
| 27 |
+
"step": 150
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.51,
|
| 31 |
+
"learning_rate": 1.6598639455782314e-05,
|
| 32 |
+
"loss": 0.0198,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.64,
|
| 37 |
+
"learning_rate": 1.5748299319727894e-05,
|
| 38 |
+
"loss": 0.0262,
|
| 39 |
+
"step": 250
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.77,
|
| 43 |
+
"learning_rate": 1.4897959183673472e-05,
|
| 44 |
+
"loss": 0.0306,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.89,
|
| 49 |
+
"learning_rate": 1.4047619047619048e-05,
|
| 50 |
+
"loss": 0.0254,
|
| 51 |
+
"step": 350
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.02,
|
| 55 |
+
"learning_rate": 1.3197278911564626e-05,
|
| 56 |
+
"loss": 0.0268,
|
| 57 |
+
"step": 400
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.15,
|
| 61 |
+
"learning_rate": 1.2346938775510204e-05,
|
| 62 |
+
"loss": 0.0018,
|
| 63 |
+
"step": 450
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.28,
|
| 67 |
+
"learning_rate": 1.1496598639455783e-05,
|
| 68 |
+
"loss": 0.0219,
|
| 69 |
+
"step": 500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.28,
|
| 73 |
+
"eval_accuracy": 0.9939432578897035,
|
| 74 |
+
"eval_loss": 0.05591470003128052,
|
| 75 |
+
"eval_runtime": 16.3109,
|
| 76 |
+
"eval_samples_per_second": 192.326,
|
| 77 |
+
"eval_steps_per_second": 12.078,
|
| 78 |
+
"step": 500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.4,
|
| 82 |
+
"learning_rate": 1.064625850340136e-05,
|
| 83 |
+
"loss": 0.0174,
|
| 84 |
+
"step": 550
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.53,
|
| 88 |
+
"learning_rate": 9.795918367346939e-06,
|
| 89 |
+
"loss": 0.0041,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.66,
|
| 94 |
+
"learning_rate": 8.945578231292518e-06,
|
| 95 |
+
"loss": 0.0101,
|
| 96 |
+
"step": 650
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.79,
|
| 100 |
+
"learning_rate": 8.095238095238097e-06,
|
| 101 |
+
"loss": 0.0102,
|
| 102 |
+
"step": 700
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.91,
|
| 106 |
+
"learning_rate": 7.244897959183675e-06,
|
| 107 |
+
"loss": 0.0003,
|
| 108 |
+
"step": 750
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.04,
|
| 112 |
+
"learning_rate": 6.394557823129253e-06,
|
| 113 |
+
"loss": 0.0055,
|
| 114 |
+
"step": 800
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.17,
|
| 118 |
+
"learning_rate": 5.54421768707483e-06,
|
| 119 |
+
"loss": 0.0004,
|
| 120 |
+
"step": 850
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.3,
|
| 124 |
+
"learning_rate": 4.693877551020409e-06,
|
| 125 |
+
"loss": 0.0016,
|
| 126 |
+
"step": 900
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.42,
|
| 130 |
+
"learning_rate": 3.843537414965986e-06,
|
| 131 |
+
"loss": 0.0001,
|
| 132 |
+
"step": 950
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.55,
|
| 136 |
+
"learning_rate": 2.993197278911565e-06,
|
| 137 |
+
"loss": 0.003,
|
| 138 |
+
"step": 1000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.55,
|
| 142 |
+
"eval_accuracy": 0.9987248963978324,
|
| 143 |
+
"eval_loss": 0.005054248962551355,
|
| 144 |
+
"eval_runtime": 16.2892,
|
| 145 |
+
"eval_samples_per_second": 192.581,
|
| 146 |
+
"eval_steps_per_second": 12.094,
|
| 147 |
+
"step": 1000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.68,
|
| 151 |
+
"learning_rate": 2.1428571428571427e-06,
|
| 152 |
+
"loss": 0.0001,
|
| 153 |
+
"step": 1050
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.81,
|
| 157 |
+
"learning_rate": 1.2925170068027212e-06,
|
| 158 |
+
"loss": 0.0053,
|
| 159 |
+
"step": 1100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.93,
|
| 163 |
+
"learning_rate": 4.421768707482994e-07,
|
| 164 |
+
"loss": 0.0001,
|
| 165 |
+
"step": 1150
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.0,
|
| 169 |
+
"step": 1176,
|
| 170 |
+
"total_flos": 7490645877017184.0,
|
| 171 |
+
"train_loss": 0.023851880903884792,
|
| 172 |
+
"train_runtime": 758.8839,
|
| 173 |
+
"train_samples_per_second": 49.517,
|
| 174 |
+
"train_steps_per_second": 1.55
|
| 175 |
+
}
|
| 176 |
+
],
|
| 177 |
+
"logging_steps": 50,
|
| 178 |
+
"max_steps": 1176,
|
| 179 |
+
"num_input_tokens_seen": 0,
|
| 180 |
+
"num_train_epochs": 3,
|
| 181 |
+
"save_steps": 500,
|
| 182 |
+
"total_flos": 7490645877017184.0,
|
| 183 |
+
"train_batch_size": 32,
|
| 184 |
+
"trial_name": null,
|
| 185 |
+
"trial_params": null
|
| 186 |
+
}
|