Model save
Browse files- trainer_state.json +86 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.142857142857142,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 80,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.1428571428571428,
|
| 13 |
+
"grad_norm": 0.9526768326759338,
|
| 14 |
+
"learning_rate": 0.0001923879532511287,
|
| 15 |
+
"loss": 1.1925,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 2.2857142857142856,
|
| 20 |
+
"grad_norm": 0.0799003466963768,
|
| 21 |
+
"learning_rate": 0.00017071067811865476,
|
| 22 |
+
"loss": 0.86,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 3.4285714285714284,
|
| 27 |
+
"grad_norm": 0.04628448933362961,
|
| 28 |
+
"learning_rate": 0.000138268343236509,
|
| 29 |
+
"loss": 0.7218,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 4.571428571428571,
|
| 34 |
+
"grad_norm": 0.04921531677246094,
|
| 35 |
+
"learning_rate": 0.0001,
|
| 36 |
+
"loss": 0.6714,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 5.714285714285714,
|
| 41 |
+
"grad_norm": 0.08176400512456894,
|
| 42 |
+
"learning_rate": 6.173165676349103e-05,
|
| 43 |
+
"loss": 0.6623,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 6.857142857142857,
|
| 48 |
+
"grad_norm": 0.056269895285367966,
|
| 49 |
+
"learning_rate": 2.9289321881345254e-05,
|
| 50 |
+
"loss": 0.6394,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 8.0,
|
| 55 |
+
"grad_norm": 0.04735175147652626,
|
| 56 |
+
"learning_rate": 7.612046748871327e-06,
|
| 57 |
+
"loss": 0.6334,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 9.142857142857142,
|
| 62 |
+
"grad_norm": 0.04113277792930603,
|
| 63 |
+
"learning_rate": 0.0,
|
| 64 |
+
"loss": 0.6326,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 9.142857142857142,
|
| 69 |
+
"step": 80,
|
| 70 |
+
"total_flos": 3.28122019872768e+16,
|
| 71 |
+
"train_loss": 0.7516828298568725,
|
| 72 |
+
"train_runtime": 7940.6726,
|
| 73 |
+
"train_samples_per_second": 0.344,
|
| 74 |
+
"train_steps_per_second": 0.01
|
| 75 |
+
}
|
| 76 |
+
],
|
| 77 |
+
"logging_steps": 10,
|
| 78 |
+
"max_steps": 80,
|
| 79 |
+
"num_input_tokens_seen": 0,
|
| 80 |
+
"num_train_epochs": 10,
|
| 81 |
+
"save_steps": 500,
|
| 82 |
+
"total_flos": 3.28122019872768e+16,
|
| 83 |
+
"train_batch_size": 8,
|
| 84 |
+
"trial_name": null,
|
| 85 |
+
"trial_params": null
|
| 86 |
+
}
|