| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.028169014084507, | |
| "eval_steps": 54, | |
| "global_step": 432, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.8726260662078857, | |
| "learning_rate": 1.267605633802817e-05, | |
| "loss": 1.2091, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.1433706283569336, | |
| "eval_runtime": 53.694, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.062, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.3356099128723145, | |
| "learning_rate": 2.535211267605634e-05, | |
| "loss": 1.0323, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.9531688690185547, | |
| "eval_runtime": 53.7017, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.061, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.767425060272217, | |
| "learning_rate": 3.802816901408451e-05, | |
| "loss": 0.9487, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.9003196954727173, | |
| "eval_runtime": 53.6993, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.061, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 2.3933067321777344, | |
| "learning_rate": 4.992175273865415e-05, | |
| "loss": 0.9267, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.8847324848175049, | |
| "eval_runtime": 53.6972, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.062, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 1.8673744201660156, | |
| "learning_rate": 4.85133020344288e-05, | |
| "loss": 0.8364, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.8761005997657776, | |
| "eval_runtime": 53.6965, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.062, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 2.8745174407958984, | |
| "learning_rate": 4.710485133020345e-05, | |
| "loss": 0.8535, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 0.8724045157432556, | |
| "eval_runtime": 53.7012, | |
| "eval_samples_per_second": 4.19, | |
| "eval_steps_per_second": 1.061, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.5754685401916504, | |
| "learning_rate": 4.569640062597809e-05, | |
| "loss": 0.8884, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 0.8635247349739075, | |
| "eval_runtime": 53.7138, | |
| "eval_samples_per_second": 4.189, | |
| "eval_steps_per_second": 1.061, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 3.002856492996216, | |
| "learning_rate": 4.428794992175274e-05, | |
| "loss": 0.8498, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.8616425395011902, | |
| "eval_runtime": 53.7191, | |
| "eval_samples_per_second": 4.188, | |
| "eval_steps_per_second": 1.061, | |
| "step": 432 | |
| } | |
| ], | |
| "logging_steps": 54, | |
| "max_steps": 2130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 54, | |
| "total_flos": 3.10783405694976e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |