Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +99 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 522,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.29,
|
| 13 |
+
"learning_rate": 1.8084291187739463e-05,
|
| 14 |
+
"loss": 1.8363,
|
| 15 |
+
"step": 50
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.57,
|
| 19 |
+
"learning_rate": 1.6168582375478928e-05,
|
| 20 |
+
"loss": 1.7907,
|
| 21 |
+
"step": 100
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.86,
|
| 25 |
+
"learning_rate": 1.4252873563218392e-05,
|
| 26 |
+
"loss": 1.7897,
|
| 27 |
+
"step": 150
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 1.15,
|
| 31 |
+
"learning_rate": 1.2337164750957855e-05,
|
| 32 |
+
"loss": 1.7892,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.44,
|
| 37 |
+
"learning_rate": 1.0421455938697318e-05,
|
| 38 |
+
"loss": 1.7858,
|
| 39 |
+
"step": 250
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.72,
|
| 43 |
+
"learning_rate": 8.505747126436782e-06,
|
| 44 |
+
"loss": 1.7702,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.01,
|
| 49 |
+
"learning_rate": 6.590038314176246e-06,
|
| 50 |
+
"loss": 1.7661,
|
| 51 |
+
"step": 350
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 2.3,
|
| 55 |
+
"learning_rate": 4.674329501915709e-06,
|
| 56 |
+
"loss": 1.7512,
|
| 57 |
+
"step": 400
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.59,
|
| 61 |
+
"learning_rate": 2.7586206896551725e-06,
|
| 62 |
+
"loss": 1.7275,
|
| 63 |
+
"step": 450
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 2.87,
|
| 67 |
+
"learning_rate": 8.429118773946361e-07,
|
| 68 |
+
"loss": 1.661,
|
| 69 |
+
"step": 500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.87,
|
| 73 |
+
"eval_accuracy": 0.3688046647230321,
|
| 74 |
+
"eval_loss": 1.633470058441162,
|
| 75 |
+
"eval_runtime": 1.6169,
|
| 76 |
+
"eval_samples_per_second": 424.268,
|
| 77 |
+
"eval_steps_per_second": 26.594,
|
| 78 |
+
"step": 500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.0,
|
| 82 |
+
"step": 522,
|
| 83 |
+
"total_flos": 1735431130580256.0,
|
| 84 |
+
"train_loss": 1.7622158079768506,
|
| 85 |
+
"train_runtime": 215.5621,
|
| 86 |
+
"train_samples_per_second": 77.212,
|
| 87 |
+
"train_steps_per_second": 2.422
|
| 88 |
+
}
|
| 89 |
+
],
|
| 90 |
+
"logging_steps": 50,
|
| 91 |
+
"max_steps": 522,
|
| 92 |
+
"num_input_tokens_seen": 0,
|
| 93 |
+
"num_train_epochs": 3,
|
| 94 |
+
"save_steps": 500,
|
| 95 |
+
"total_flos": 1735431130580256.0,
|
| 96 |
+
"train_batch_size": 32,
|
| 97 |
+
"trial_name": null,
|
| 98 |
+
"trial_params": null
|
| 99 |
+
}
|