|
{ |
|
"best_metric": 0.2654401361942291, |
|
"best_model_checkpoint": "/kaggle/working/results/checkpoint-170", |
|
"epoch": 5.0, |
|
"eval_steps": 500, |
|
"global_step": 170, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.29411764705882354, |
|
"grad_norm": 1.7904845476150513, |
|
"learning_rate": 4.705882352941177e-05, |
|
"loss": 1.3311, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.5882352941176471, |
|
"grad_norm": 1.4475995302200317, |
|
"learning_rate": 4.411764705882353e-05, |
|
"loss": 0.7478, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.8823529411764706, |
|
"grad_norm": 1.0543155670166016, |
|
"learning_rate": 4.11764705882353e-05, |
|
"loss": 0.5724, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9163346613545816, |
|
"eval_loss": 0.42589500546455383, |
|
"eval_runtime": 5.1988, |
|
"eval_samples_per_second": 48.28, |
|
"eval_steps_per_second": 0.769, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 1.1764705882352942, |
|
"grad_norm": 0.9727240204811096, |
|
"learning_rate": 3.8235294117647055e-05, |
|
"loss": 0.4738, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.4705882352941178, |
|
"grad_norm": 0.6961530447006226, |
|
"learning_rate": 3.529411764705883e-05, |
|
"loss": 0.4334, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.7647058823529411, |
|
"grad_norm": 0.9150127172470093, |
|
"learning_rate": 3.235294117647059e-05, |
|
"loss": 0.3558, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9362549800796812, |
|
"eval_loss": 0.3115582764148712, |
|
"eval_runtime": 5.2552, |
|
"eval_samples_per_second": 47.762, |
|
"eval_steps_per_second": 0.761, |
|
"step": 68 |
|
}, |
|
{ |
|
"epoch": 2.0588235294117645, |
|
"grad_norm": 0.8191739916801453, |
|
"learning_rate": 2.9411764705882354e-05, |
|
"loss": 0.3313, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.3529411764705883, |
|
"grad_norm": 0.8822425007820129, |
|
"learning_rate": 2.647058823529412e-05, |
|
"loss": 0.3025, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 2.6470588235294117, |
|
"grad_norm": 1.1780915260314941, |
|
"learning_rate": 2.3529411764705884e-05, |
|
"loss": 0.2882, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 2.9411764705882355, |
|
"grad_norm": 1.081735610961914, |
|
"learning_rate": 2.058823529411765e-05, |
|
"loss": 0.2732, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9362549800796812, |
|
"eval_loss": 0.28424072265625, |
|
"eval_runtime": 5.1718, |
|
"eval_samples_per_second": 48.532, |
|
"eval_steps_per_second": 0.773, |
|
"step": 102 |
|
}, |
|
{ |
|
"epoch": 3.235294117647059, |
|
"grad_norm": 0.8298177123069763, |
|
"learning_rate": 1.7647058823529414e-05, |
|
"loss": 0.2513, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 3.5294117647058822, |
|
"grad_norm": 0.933785080909729, |
|
"learning_rate": 1.4705882352941177e-05, |
|
"loss": 0.2497, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 3.8235294117647056, |
|
"grad_norm": 0.7405449748039246, |
|
"learning_rate": 1.1764705882352942e-05, |
|
"loss": 0.2286, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9402390438247012, |
|
"eval_loss": 0.269024133682251, |
|
"eval_runtime": 5.2408, |
|
"eval_samples_per_second": 47.894, |
|
"eval_steps_per_second": 0.763, |
|
"step": 136 |
|
}, |
|
{ |
|
"epoch": 4.117647058823529, |
|
"grad_norm": 0.9281182289123535, |
|
"learning_rate": 8.823529411764707e-06, |
|
"loss": 0.2377, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 4.411764705882353, |
|
"grad_norm": 1.4719483852386475, |
|
"learning_rate": 5.882352941176471e-06, |
|
"loss": 0.2671, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 4.705882352941177, |
|
"grad_norm": 0.7358911633491516, |
|
"learning_rate": 2.9411764705882355e-06, |
|
"loss": 0.218, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"grad_norm": 1.2096456289291382, |
|
"learning_rate": 0.0, |
|
"loss": 0.1984, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9402390438247012, |
|
"eval_loss": 0.2654401361942291, |
|
"eval_runtime": 5.4015, |
|
"eval_samples_per_second": 46.468, |
|
"eval_steps_per_second": 0.741, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"step": 170, |
|
"total_flos": 8.264816981535744e+17, |
|
"train_loss": 0.3976527817109052, |
|
"train_runtime": 377.5007, |
|
"train_samples_per_second": 28.252, |
|
"train_steps_per_second": 0.45 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 170, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 8.264816981535744e+17, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|