| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 50, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2, |
| "grad_norm": 18.864488225514396, |
| "learning_rate": 2.0000000000000003e-06, |
| "loss": 0.9524, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 16.624666091321274, |
| "learning_rate": 1e-05, |
| "loss": 0.8684, |
| "step": 5 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.7913689017295837, |
| "eval_runtime": 5.7172, |
| "eval_samples_per_second": 54.397, |
| "eval_steps_per_second": 1.749, |
| "step": 5 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 4.401290119342954, |
| "learning_rate": 9.698463103929542e-06, |
| "loss": 0.7031, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.5348705053329468, |
| "eval_runtime": 5.7154, |
| "eval_samples_per_second": 54.414, |
| "eval_steps_per_second": 1.75, |
| "step": 10 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.029002967994018, |
| "learning_rate": 8.83022221559489e-06, |
| "loss": 0.4888, |
| "step": 15 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.34116050601005554, |
| "eval_runtime": 5.7277, |
| "eval_samples_per_second": 54.298, |
| "eval_steps_per_second": 1.746, |
| "step": 15 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.1549630858744604, |
| "learning_rate": 7.500000000000001e-06, |
| "loss": 0.302, |
| "step": 20 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.19024085998535156, |
| "eval_runtime": 5.7234, |
| "eval_samples_per_second": 54.338, |
| "eval_steps_per_second": 1.747, |
| "step": 20 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 4.126405291248831, |
| "learning_rate": 5.8682408883346535e-06, |
| "loss": 0.1643, |
| "step": 25 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.08896404504776001, |
| "eval_runtime": 5.7273, |
| "eval_samples_per_second": 54.301, |
| "eval_steps_per_second": 1.746, |
| "step": 25 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.870513503322164, |
| "learning_rate": 4.131759111665349e-06, |
| "loss": 0.0778, |
| "step": 30 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.04157733544707298, |
| "eval_runtime": 5.7247, |
| "eval_samples_per_second": 54.326, |
| "eval_steps_per_second": 1.747, |
| "step": 30 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.7040194155687554, |
| "learning_rate": 2.5000000000000015e-06, |
| "loss": 0.0404, |
| "step": 35 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.027967050671577454, |
| "eval_runtime": 5.7274, |
| "eval_samples_per_second": 54.301, |
| "eval_steps_per_second": 1.746, |
| "step": 35 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 1.2638859662186757, |
| "learning_rate": 1.1697777844051105e-06, |
| "loss": 0.0279, |
| "step": 40 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.021927356719970703, |
| "eval_runtime": 5.7381, |
| "eval_samples_per_second": 54.199, |
| "eval_steps_per_second": 1.743, |
| "step": 40 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.9658748602996367, |
| "learning_rate": 3.015368960704584e-07, |
| "loss": 0.0214, |
| "step": 45 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.018554512411355972, |
| "eval_runtime": 5.7334, |
| "eval_samples_per_second": 54.244, |
| "eval_steps_per_second": 1.744, |
| "step": 45 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.5265240066937839, |
| "learning_rate": 0.0, |
| "loss": 0.0183, |
| "step": 50 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.017760511487722397, |
| "eval_runtime": 5.6974, |
| "eval_samples_per_second": 54.586, |
| "eval_steps_per_second": 1.755, |
| "step": 50 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 50, |
| "total_flos": 10468982784000.0, |
| "train_loss": 0.2729184678196907, |
| "train_runtime": 352.114, |
| "train_samples_per_second": 8.832, |
| "train_steps_per_second": 0.142 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 50, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 10468982784000.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
