|
{ |
|
"best_metric": 0.4741882085800171, |
|
"best_model_checkpoint": "mikhail-panzo/fil_b32_le3_s4000/checkpoint-500", |
|
"epoch": 21.978021978021978, |
|
"eval_steps": 500, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.098901098901099, |
|
"grad_norm": 2.8254339694976807, |
|
"learning_rate": 2.5e-05, |
|
"loss": 0.736, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.197802197802198, |
|
"grad_norm": 4.995375633239746, |
|
"learning_rate": 4.9500000000000004e-05, |
|
"loss": 0.5913, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 3.2967032967032965, |
|
"grad_norm": 2.352879285812378, |
|
"learning_rate": 7.45e-05, |
|
"loss": 0.5174, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 4.395604395604396, |
|
"grad_norm": 4.0169596672058105, |
|
"learning_rate": 9.95e-05, |
|
"loss": 0.5275, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 5.4945054945054945, |
|
"grad_norm": 5.557097434997559, |
|
"learning_rate": 0.0001245, |
|
"loss": 0.5231, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 6.593406593406593, |
|
"grad_norm": 8.755484580993652, |
|
"learning_rate": 0.0001495, |
|
"loss": 0.5127, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 7.6923076923076925, |
|
"grad_norm": 2.5811164379119873, |
|
"learning_rate": 0.00017449999999999999, |
|
"loss": 0.5179, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 8.791208791208792, |
|
"grad_norm": 7.064955234527588, |
|
"learning_rate": 0.00019950000000000002, |
|
"loss": 0.4961, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 9.89010989010989, |
|
"grad_norm": 8.559194564819336, |
|
"learning_rate": 0.0002245, |
|
"loss": 0.4976, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 10.989010989010989, |
|
"grad_norm": 7.322611331939697, |
|
"learning_rate": 0.0002495, |
|
"loss": 0.502, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 10.989010989010989, |
|
"eval_loss": 0.4741882085800171, |
|
"eval_runtime": 7.5067, |
|
"eval_samples_per_second": 21.448, |
|
"eval_steps_per_second": 2.798, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 12.087912087912088, |
|
"grad_norm": 3.533902883529663, |
|
"learning_rate": 0.0002745, |
|
"loss": 0.5146, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 13.186813186813186, |
|
"grad_norm": 9.026822090148926, |
|
"learning_rate": 0.0002995, |
|
"loss": 0.4956, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 14.285714285714286, |
|
"grad_norm": 7.3526291847229, |
|
"learning_rate": 0.00032450000000000003, |
|
"loss": 0.5406, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 15.384615384615385, |
|
"grad_norm": 2.889016628265381, |
|
"learning_rate": 0.0003495, |
|
"loss": 0.5153, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 16.483516483516482, |
|
"grad_norm": 7.851646423339844, |
|
"learning_rate": 0.0003745, |
|
"loss": 0.5326, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 17.582417582417584, |
|
"grad_norm": 2.468174934387207, |
|
"learning_rate": 0.0003995, |
|
"loss": 0.5575, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 18.681318681318682, |
|
"grad_norm": 10.550352096557617, |
|
"learning_rate": 0.0004245, |
|
"loss": 0.5893, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 19.78021978021978, |
|
"grad_norm": 6.15378475189209, |
|
"learning_rate": 0.00044950000000000003, |
|
"loss": 0.5769, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 20.87912087912088, |
|
"grad_norm": 8.328354835510254, |
|
"learning_rate": 0.0004745, |
|
"loss": 0.541, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 21.978021978021978, |
|
"grad_norm": 6.08469820022583, |
|
"learning_rate": 0.0004995, |
|
"loss": 0.5446, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 21.978021978021978, |
|
"eval_loss": 0.4899168014526367, |
|
"eval_runtime": 6.7313, |
|
"eval_samples_per_second": 23.918, |
|
"eval_steps_per_second": 3.12, |
|
"step": 1000 |
|
} |
|
], |
|
"logging_steps": 50, |
|
"max_steps": 4000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 89, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 7353193286254464.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|