| { | |
| "best_metric": 1.0291192531585693, | |
| "best_model_checkpoint": "savedEpoch_nccl/checkpoint-91656", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 91656, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.6372342109680176, | |
| "learning_rate": 9.882461854198269e-05, | |
| "loss": 1.1441, | |
| "step": 15276 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0554957389831543, | |
| "eval_runtime": 57.5344, | |
| "eval_samples_per_second": 12743.952, | |
| "eval_steps_per_second": 33.198, | |
| "step": 15276 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6399237513542175, | |
| "learning_rate": 9.541848844472206e-05, | |
| "loss": 1.0486, | |
| "step": 30552 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0447489023208618, | |
| "eval_runtime": 240.72, | |
| "eval_samples_per_second": 3045.924, | |
| "eval_steps_per_second": 7.935, | |
| "step": 30552 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.7907009124755859, | |
| "learning_rate": 9.201235834746142e-05, | |
| "loss": 1.0407, | |
| "step": 45828 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.0373780727386475, | |
| "eval_runtime": 56.906, | |
| "eval_samples_per_second": 12884.672, | |
| "eval_steps_per_second": 33.564, | |
| "step": 45828 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5956621766090393, | |
| "learning_rate": 8.860622825020078e-05, | |
| "loss": 1.0376, | |
| "step": 61104 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.0327318906784058, | |
| "eval_runtime": 57.0015, | |
| "eval_samples_per_second": 12863.086, | |
| "eval_steps_per_second": 33.508, | |
| "step": 61104 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6499078273773193, | |
| "learning_rate": 8.519965200321228e-05, | |
| "loss": 1.037, | |
| "step": 76380 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.0331288576126099, | |
| "eval_runtime": 57.0146, | |
| "eval_samples_per_second": 12860.131, | |
| "eval_steps_per_second": 33.5, | |
| "step": 76380 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.5220733284950256, | |
| "learning_rate": 8.179329883108772e-05, | |
| "loss": 1.0344, | |
| "step": 91656 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.0291192531585693, | |
| "eval_runtime": 56.9287, | |
| "eval_samples_per_second": 12879.541, | |
| "eval_steps_per_second": 33.551, | |
| "step": 91656 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 458280, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 8.139053392031908e+17, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |