| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 59, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 4.72473337945939, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.9545, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 15.606619949426117, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.9459, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 97.5546177933576, | |
| "learning_rate": 1.972022914080411e-05, | |
| "loss": 1.309, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 7.451729536881689, | |
| "learning_rate": 1.8610436117673557e-05, | |
| "loss": 1.1022, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 4.013413307542302, | |
| "learning_rate": 1.6749830015182106e-05, | |
| "loss": 0.9536, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 2.5281389594304153, | |
| "learning_rate": 1.4300652022765207e-05, | |
| "loss": 0.9122, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.4087072108196184, | |
| "learning_rate": 1.1476465640024814e-05, | |
| "loss": 0.8548, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.127180353057553, | |
| "learning_rate": 8.52353435997519e-06, | |
| "loss": 0.8386, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.9083395084428687, | |
| "learning_rate": 5.699347977234799e-06, | |
| "loss": 0.8324, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.7712149272954687, | |
| "learning_rate": 3.250169984817897e-06, | |
| "loss": 0.7792, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.6965019256204015, | |
| "learning_rate": 1.3895638823264447e-06, | |
| "loss": 0.77, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.6261522519462899, | |
| "learning_rate": 2.7977085919589253e-07, | |
| "loss": 0.7609, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7076106071472168, | |
| "eval_runtime": 338.673, | |
| "eval_samples_per_second": 22.207, | |
| "eval_steps_per_second": 0.348, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 59, | |
| "total_flos": 49413598740480.0, | |
| "train_loss": 0.908859632783017, | |
| "train_runtime": 1626.0912, | |
| "train_samples_per_second": 4.625, | |
| "train_steps_per_second": 0.036 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 59, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 49413598740480.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |