| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8860759493670884, | |
| "eval_steps": 500, | |
| "global_step": 57, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.2923335171622337, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 2.1896, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.9912276172628393, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.3641, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.815034805697151, | |
| "learning_rate": 1.9697969360350098e-05, | |
| "loss": 2.4061, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.6786695631350969, | |
| "learning_rate": 1.8502171357296144e-05, | |
| "loss": 2.4036, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.4486770629882812, | |
| "eval_runtime": 1.5127, | |
| "eval_samples_per_second": 5.289, | |
| "eval_steps_per_second": 0.661, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.6621397636588793, | |
| "learning_rate": 1.650618300204242e-05, | |
| "loss": 2.4408, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 0.8896671084173804, | |
| "learning_rate": 1.3897858732926794e-05, | |
| "loss": 2.1577, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.7473486744572, | |
| "learning_rate": 1.092268359463302e-05, | |
| "loss": 1.9602, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.7400600277246799, | |
| "learning_rate": 7.860669167935028e-06, | |
| "loss": 2.0418, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.485201358795166, | |
| "eval_runtime": 1.4984, | |
| "eval_samples_per_second": 5.339, | |
| "eval_steps_per_second": 0.667, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.8160514109181762, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 2.0077, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 0.7939146315083865, | |
| "learning_rate": 2.6099108277934105e-06, | |
| "loss": 1.7591, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.8611599565796522, | |
| "learning_rate": 9.153472818047627e-07, | |
| "loss": 1.8598, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.809915034715595, | |
| "learning_rate": 7.579490328064265e-08, | |
| "loss": 1.8471, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_loss": 1.5596716403961182, | |
| "eval_runtime": 1.7332, | |
| "eval_samples_per_second": 4.616, | |
| "eval_steps_per_second": 0.577, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "step": 57, | |
| "total_flos": 11882295459840.0, | |
| "train_loss": 2.103332046876874, | |
| "train_runtime": 1101.4918, | |
| "train_samples_per_second": 3.418, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 57, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 11882295459840.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |