Commit
•
bf47ed6
1
Parent(s):
ffd5579
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +133 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,133 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7647,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2,
|
| 13 |
+
"learning_rate": 2.8038446449588075e-05,
|
| 14 |
+
"loss": 2.5909,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.39,
|
| 19 |
+
"learning_rate": 2.6076892899176145e-05,
|
| 20 |
+
"loss": 2.3461,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.59,
|
| 25 |
+
"learning_rate": 2.4115339348764223e-05,
|
| 26 |
+
"loss": 2.2111,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.78,
|
| 31 |
+
"learning_rate": 2.2153785798352297e-05,
|
| 32 |
+
"loss": 2.1523,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.98,
|
| 37 |
+
"learning_rate": 2.019223224794037e-05,
|
| 38 |
+
"loss": 2.0715,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0,
|
| 43 |
+
"eval_loss": 1.9488946199417114,
|
| 44 |
+
"eval_runtime": 854.0815,
|
| 45 |
+
"eval_samples_per_second": 2.984,
|
| 46 |
+
"eval_steps_per_second": 0.747,
|
| 47 |
+
"step": 2549
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.18,
|
| 51 |
+
"learning_rate": 1.8230678697528442e-05,
|
| 52 |
+
"loss": 1.4507,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.37,
|
| 57 |
+
"learning_rate": 1.6269125147116516e-05,
|
| 58 |
+
"loss": 1.3564,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.57,
|
| 63 |
+
"learning_rate": 1.430757159670459e-05,
|
| 64 |
+
"loss": 1.3545,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.77,
|
| 69 |
+
"learning_rate": 1.2346018046292664e-05,
|
| 70 |
+
"loss": 1.3372,
|
| 71 |
+
"step": 4500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.96,
|
| 75 |
+
"learning_rate": 1.0384464495880738e-05,
|
| 76 |
+
"loss": 1.3039,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.0,
|
| 81 |
+
"eval_loss": 1.8286019563674927,
|
| 82 |
+
"eval_runtime": 849.7166,
|
| 83 |
+
"eval_samples_per_second": 3.0,
|
| 84 |
+
"eval_steps_per_second": 0.751,
|
| 85 |
+
"step": 5098
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 2.16,
|
| 89 |
+
"learning_rate": 8.42291094546881e-06,
|
| 90 |
+
"loss": 0.987,
|
| 91 |
+
"step": 5500
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 2.35,
|
| 95 |
+
"learning_rate": 6.461357395056885e-06,
|
| 96 |
+
"loss": 0.8986,
|
| 97 |
+
"step": 6000
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.55,
|
| 101 |
+
"learning_rate": 4.4998038446449585e-06,
|
| 102 |
+
"loss": 0.9008,
|
| 103 |
+
"step": 6500
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 2.75,
|
| 107 |
+
"learning_rate": 2.538250294233033e-06,
|
| 108 |
+
"loss": 0.8903,
|
| 109 |
+
"step": 7000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.94,
|
| 113 |
+
"learning_rate": 5.766967438211064e-07,
|
| 114 |
+
"loss": 0.8313,
|
| 115 |
+
"step": 7500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 3.0,
|
| 119 |
+
"eval_loss": 1.8424451351165771,
|
| 120 |
+
"eval_runtime": 846.7842,
|
| 121 |
+
"eval_samples_per_second": 3.01,
|
| 122 |
+
"eval_steps_per_second": 0.753,
|
| 123 |
+
"step": 7647
|
| 124 |
+
}
|
| 125 |
+
],
|
| 126 |
+
"logging_steps": 500,
|
| 127 |
+
"max_steps": 7647,
|
| 128 |
+
"num_train_epochs": 3,
|
| 129 |
+
"save_steps": 500,
|
| 130 |
+
"total_flos": 1.6566780520562688e+16,
|
| 131 |
+
"trial_name": null,
|
| 132 |
+
"trial_params": null
|
| 133 |
+
}
|