| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.42216358839050133, |
| "eval_steps": 12, |
| "global_step": 20, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0, |
| "eval_loss": 1.051796555519104, |
| "eval_ppl": 2.86279, |
| "eval_runtime": 277.2697, |
| "eval_samples_per_second": 0.573, |
| "eval_steps_per_second": 0.144, |
| "memory/device_reserved (GiB)": 45.79, |
| "memory/max_active (GiB)": 41.03, |
| "memory/max_allocated (GiB)": 38.11, |
| "step": 0 |
| }, |
| { |
| "epoch": 0.10554089709762533, |
| "grad_norm": 3.6152215003967285, |
| "learning_rate": 7e-06, |
| "loss": 0.9618721008300781, |
| "memory/device_reserved (GiB)": 76.43, |
| "memory/max_active (GiB)": 61.19, |
| "memory/max_allocated (GiB)": 58.82, |
| "ppl": 2.61659, |
| "step": 5, |
| "tokens/total": 2621440, |
| "tokens/trainable": 439427 |
| }, |
| { |
| "epoch": 0.21108179419525067, |
| "grad_norm": 2.375864267349243, |
| "learning_rate": 6.952144421988455e-06, |
| "loss": 0.7351778507232666, |
| "memory/device_reserved (GiB)": 76.43, |
| "memory/max_active (GiB)": 61.19, |
| "memory/max_allocated (GiB)": 58.82, |
| "ppl": 2.08585, |
| "step": 10, |
| "tokens/total": 5242880, |
| "tokens/train_per_sec_per_gpu": 53.56, |
| "tokens/trainable": 905507 |
| }, |
| { |
| "epoch": 0.2532981530343008, |
| "eval_loss": 0.6750471591949463, |
| "eval_ppl": 1.96413, |
| "eval_runtime": 269.2239, |
| "eval_samples_per_second": 0.591, |
| "eval_steps_per_second": 0.149, |
| "memory/device_reserved (GiB)": 76.43, |
| "memory/max_active (GiB)": 61.19, |
| "memory/max_allocated (GiB)": 58.82, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.316622691292876, |
| "grad_norm": 2.0795094966888428, |
| "learning_rate": 6.810031755475611e-06, |
| "loss": 0.6558361530303956, |
| "memory/device_reserved (GiB)": 76.89, |
| "memory/max_active (GiB)": 61.19, |
| "memory/max_allocated (GiB)": 58.82, |
| "ppl": 1.92675, |
| "step": 15, |
| "tokens/total": 7864320, |
| "tokens/train_per_sec_per_gpu": 31.8, |
| "tokens/trainable": 1335905 |
| }, |
| { |
| "epoch": 0.42216358839050133, |
| "grad_norm": 1.7829387187957764, |
| "learning_rate": 6.577980021920982e-06, |
| "loss": 0.5943492412567138, |
| "memory/device_reserved (GiB)": 76.74, |
| "memory/max_active (GiB)": 61.19, |
| "memory/max_allocated (GiB)": 58.82, |
| "ppl": 1.81185, |
| "step": 20, |
| "tokens/total": 10485760, |
| "tokens/train_per_sec_per_gpu": 40.47, |
| "tokens/trainable": 1759194 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 94, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 10, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 190536031928320.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
