Commit
•
3e1c70f
1
Parent(s):
c240149
Upload trainer_state.json
Browse files- trainer_state.json +184 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,184 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9087947882736156,
|
| 5 |
+
"global_step": 12000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 0.0019185667752442998,
|
| 13 |
+
"loss": 2.9941,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.33,
|
| 18 |
+
"learning_rate": 0.0018371335504885993,
|
| 19 |
+
"loss": 1.5024,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.49,
|
| 24 |
+
"learning_rate": 0.001755700325732899,
|
| 25 |
+
"loss": 0.0902,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.65,
|
| 30 |
+
"learning_rate": 0.0016742671009771987,
|
| 31 |
+
"loss": 0.0504,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.81,
|
| 36 |
+
"learning_rate": 0.0015928338762214984,
|
| 37 |
+
"loss": 0.0412,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.98,
|
| 42 |
+
"learning_rate": 0.0015114006514657982,
|
| 43 |
+
"loss": 0.0373,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_loss": 0.026404375210404396,
|
| 49 |
+
"eval_runtime": 68.798,
|
| 50 |
+
"eval_samples_per_second": 178.45,
|
| 51 |
+
"eval_steps_per_second": 11.163,
|
| 52 |
+
"step": 3070
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.14,
|
| 56 |
+
"learning_rate": 0.0014299674267100977,
|
| 57 |
+
"loss": 0.0346,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.3,
|
| 62 |
+
"learning_rate": 0.0013485342019543974,
|
| 63 |
+
"loss": 0.0301,
|
| 64 |
+
"step": 4000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 1.47,
|
| 68 |
+
"learning_rate": 0.001267100977198697,
|
| 69 |
+
"loss": 0.0323,
|
| 70 |
+
"step": 4500
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 1.63,
|
| 74 |
+
"learning_rate": 0.0011856677524429969,
|
| 75 |
+
"loss": 0.0305,
|
| 76 |
+
"step": 5000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.79,
|
| 80 |
+
"learning_rate": 0.0011042345276872966,
|
| 81 |
+
"loss": 0.0287,
|
| 82 |
+
"step": 5500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.95,
|
| 86 |
+
"learning_rate": 0.001022801302931596,
|
| 87 |
+
"loss": 0.0287,
|
| 88 |
+
"step": 6000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.0,
|
| 92 |
+
"eval_loss": 0.023602882400155067,
|
| 93 |
+
"eval_runtime": 68.7368,
|
| 94 |
+
"eval_samples_per_second": 178.609,
|
| 95 |
+
"eval_steps_per_second": 11.173,
|
| 96 |
+
"step": 6140
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 2.12,
|
| 100 |
+
"learning_rate": 0.0009413680781758957,
|
| 101 |
+
"loss": 0.0261,
|
| 102 |
+
"step": 6500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.28,
|
| 106 |
+
"learning_rate": 0.0008599348534201955,
|
| 107 |
+
"loss": 0.0255,
|
| 108 |
+
"step": 7000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.44,
|
| 112 |
+
"learning_rate": 0.0007785016286644952,
|
| 113 |
+
"loss": 0.025,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.61,
|
| 118 |
+
"learning_rate": 0.0006970684039087948,
|
| 119 |
+
"loss": 0.0247,
|
| 120 |
+
"step": 8000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.77,
|
| 124 |
+
"learning_rate": 0.0006156351791530945,
|
| 125 |
+
"loss": 0.023,
|
| 126 |
+
"step": 8500
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.93,
|
| 130 |
+
"learning_rate": 0.0005342019543973941,
|
| 131 |
+
"loss": 0.0218,
|
| 132 |
+
"step": 9000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 3.0,
|
| 136 |
+
"eval_loss": 0.02037286013364792,
|
| 137 |
+
"eval_runtime": 68.7895,
|
| 138 |
+
"eval_samples_per_second": 178.472,
|
| 139 |
+
"eval_steps_per_second": 11.165,
|
| 140 |
+
"step": 9210
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.09,
|
| 144 |
+
"learning_rate": 0.0004527687296416938,
|
| 145 |
+
"loss": 0.0198,
|
| 146 |
+
"step": 9500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.26,
|
| 150 |
+
"learning_rate": 0.0003713355048859935,
|
| 151 |
+
"loss": 0.0187,
|
| 152 |
+
"step": 10000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.42,
|
| 156 |
+
"learning_rate": 0.0002899022801302932,
|
| 157 |
+
"loss": 0.0181,
|
| 158 |
+
"step": 10500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.58,
|
| 162 |
+
"learning_rate": 0.00020846905537459285,
|
| 163 |
+
"loss": 0.0174,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.75,
|
| 168 |
+
"learning_rate": 0.00012703583061889252,
|
| 169 |
+
"loss": 0.0166,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.91,
|
| 174 |
+
"learning_rate": 4.5602605863192184e-05,
|
| 175 |
+
"loss": 0.0167,
|
| 176 |
+
"step": 12000
|
| 177 |
+
}
|
| 178 |
+
],
|
| 179 |
+
"max_steps": 12280,
|
| 180 |
+
"num_train_epochs": 4,
|
| 181 |
+
"total_flos": 2.5979941526962176e+16,
|
| 182 |
+
"trial_name": null,
|
| 183 |
+
"trial_params": null
|
| 184 |
+
}
|