Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +106 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,106 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 3.054584503173828,
|
| 3 |
+
"best_model_checkpoint": "./results/checkpoint-750",
|
| 4 |
+
"epoch": 0.3634161114476075,
|
| 5 |
+
"eval_steps": 250,
|
| 6 |
+
"global_step": 750,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04845548152634767,
|
| 13 |
+
"grad_norm": 0.2021484375,
|
| 14 |
+
"learning_rate": 0.0019926230341909047,
|
| 15 |
+
"loss": 5.5548,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09691096305269534,
|
| 20 |
+
"grad_norm": 0.484375,
|
| 21 |
+
"learning_rate": 0.0019623193935821215,
|
| 22 |
+
"loss": 4.0801,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12113870381586916,
|
| 27 |
+
"eval_loss": 3.651378631591797,
|
| 28 |
+
"eval_runtime": 1225.4443,
|
| 29 |
+
"eval_samples_per_second": 47.902,
|
| 30 |
+
"eval_steps_per_second": 1.497,
|
| 31 |
+
"step": 250
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.145366444579043,
|
| 35 |
+
"grad_norm": 0.6796875,
|
| 36 |
+
"learning_rate": 0.0019092830690827923,
|
| 37 |
+
"loss": 3.6511,
|
| 38 |
+
"step": 300
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.19382192610539067,
|
| 42 |
+
"grad_norm": 0.474609375,
|
| 43 |
+
"learning_rate": 0.0018347669274724923,
|
| 44 |
+
"loss": 3.4945,
|
| 45 |
+
"step": 400
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.24227740763173833,
|
| 49 |
+
"grad_norm": 0.51171875,
|
| 50 |
+
"learning_rate": 0.0017405312490731885,
|
| 51 |
+
"loss": 3.3662,
|
| 52 |
+
"step": 500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.24227740763173833,
|
| 56 |
+
"eval_loss": 3.2901480197906494,
|
| 57 |
+
"eval_runtime": 1223.1792,
|
| 58 |
+
"eval_samples_per_second": 47.991,
|
| 59 |
+
"eval_steps_per_second": 1.5,
|
| 60 |
+
"step": 500
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.290732889158086,
|
| 64 |
+
"grad_norm": 0.90234375,
|
| 65 |
+
"learning_rate": 0.0016288021449895208,
|
| 66 |
+
"loss": 3.2248,
|
| 67 |
+
"step": 600
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.3391883706844337,
|
| 71 |
+
"grad_norm": 0.5859375,
|
| 72 |
+
"learning_rate": 0.001502218970119089,
|
| 73 |
+
"loss": 3.1488,
|
| 74 |
+
"step": 700
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3634161114476075,
|
| 78 |
+
"eval_loss": 3.054584503173828,
|
| 79 |
+
"eval_runtime": 1225.6755,
|
| 80 |
+
"eval_samples_per_second": 47.893,
|
| 81 |
+
"eval_steps_per_second": 1.497,
|
| 82 |
+
"step": 750
|
| 83 |
+
}
|
| 84 |
+
],
|
| 85 |
+
"logging_steps": 100,
|
| 86 |
+
"max_steps": 2063,
|
| 87 |
+
"num_input_tokens_seen": 0,
|
| 88 |
+
"num_train_epochs": 1,
|
| 89 |
+
"save_steps": 250,
|
| 90 |
+
"stateful_callbacks": {
|
| 91 |
+
"TrainerControl": {
|
| 92 |
+
"args": {
|
| 93 |
+
"should_epoch_stop": false,
|
| 94 |
+
"should_evaluate": false,
|
| 95 |
+
"should_log": false,
|
| 96 |
+
"should_save": true,
|
| 97 |
+
"should_training_stop": false
|
| 98 |
+
},
|
| 99 |
+
"attributes": {}
|
| 100 |
+
}
|
| 101 |
+
},
|
| 102 |
+
"total_flos": 1.754471054966784e+17,
|
| 103 |
+
"train_batch_size": 4,
|
| 104 |
+
"trial_name": null,
|
| 105 |
+
"trial_params": null
|
| 106 |
+
}
|