| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 4500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 10.763190269470215, | |
| "learning_rate": 3.335112059765208e-05, | |
| "loss": 1.5382, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 4.8812, | |
| "eval_loss": 1.1502126455307007, | |
| "eval_rouge1": 60.247880952380996, | |
| "eval_rouge2": 42.58590476190485, | |
| "eval_rougeL": 60.23339682539681, | |
| "eval_rougeLsum": 60.23446031746034, | |
| "eval_runtime": 1016.7664, | |
| "eval_samples_per_second": 4.918, | |
| "eval_steps_per_second": 1.229, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 4.3876237869262695, | |
| "learning_rate": 6.670224119530416e-06, | |
| "loss": 1.2867, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.068687915802002, | |
| "learning_rate": 0.0, | |
| "loss": 1.1906, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 4.9042, | |
| "eval_loss": 1.1154725551605225, | |
| "eval_rouge1": 60.75021428571432, | |
| "eval_rouge2": 43.73785714285723, | |
| "eval_rougeL": 60.75853968253971, | |
| "eval_rougeLsum": 60.7197619047618, | |
| "eval_runtime": 1019.2932, | |
| "eval_samples_per_second": 4.905, | |
| "eval_steps_per_second": 1.226, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 6.656228065490723, | |
| "learning_rate": 0.0, | |
| "loss": 1.2365, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 4.9042, | |
| "eval_loss": 1.1154857873916626, | |
| "eval_rouge1": 60.75021428571432, | |
| "eval_rouge2": 43.73785714285723, | |
| "eval_rougeL": 60.75853968253971, | |
| "eval_rougeLsum": 60.7197619047618, | |
| "eval_runtime": 1019.9367, | |
| "eval_samples_per_second": 4.902, | |
| "eval_steps_per_second": 1.226, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 8.076435089111328, | |
| "learning_rate": 0.0, | |
| "loss": 1.2041, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.218427896499634, | |
| "learning_rate": 0.0, | |
| "loss": 1.1962, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 4.9042, | |
| "eval_loss": 1.1154814958572388, | |
| "eval_rouge1": 60.75021428571432, | |
| "eval_rouge2": 43.73785714285723, | |
| "eval_rougeL": 60.75853968253971, | |
| "eval_rougeLsum": 60.7197619047618, | |
| "eval_runtime": 1019.891, | |
| "eval_samples_per_second": 4.902, | |
| "eval_steps_per_second": 1.226, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "grad_norm": 3.1441781520843506, | |
| "learning_rate": 0.0, | |
| "loss": 1.2323, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 4.9042, | |
| "eval_loss": 1.1154634952545166, | |
| "eval_rouge1": 60.75021428571432, | |
| "eval_rouge2": 43.73785714285723, | |
| "eval_rougeL": 60.75853968253971, | |
| "eval_rougeLsum": 60.7197619047618, | |
| "eval_runtime": 1016.3103, | |
| "eval_samples_per_second": 4.92, | |
| "eval_steps_per_second": 1.23, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 4.466716766357422, | |
| "learning_rate": 0.0, | |
| "loss": 1.2174, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 5.4029083251953125, | |
| "learning_rate": 0.0, | |
| "loss": 1.1927, | |
| "step": 4500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.1059086753792e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |