| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.788732394366197, | |
| "eval_steps": 54, | |
| "global_step": 594, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.0718016624450684, | |
| "learning_rate": 1.267605633802817e-05, | |
| "loss": 1.0998, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.0505307912826538, | |
| "eval_runtime": 60.3266, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 3.964005947113037, | |
| "learning_rate": 2.535211267605634e-05, | |
| "loss": 0.9713, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.8683193922042847, | |
| "eval_runtime": 60.3172, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.101376533508301, | |
| "learning_rate": 3.802816901408451e-05, | |
| "loss": 0.8291, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.810826301574707, | |
| "eval_runtime": 60.3149, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 1.6203935146331787, | |
| "learning_rate": 4.992175273865415e-05, | |
| "loss": 0.7965, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.784785807132721, | |
| "eval_runtime": 60.3247, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.086142063140869, | |
| "learning_rate": 4.85133020344288e-05, | |
| "loss": 0.7287, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.7797535061836243, | |
| "eval_runtime": 60.326, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.9750502109527588, | |
| "learning_rate": 4.710485133020345e-05, | |
| "loss": 0.7509, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 0.7676067352294922, | |
| "eval_runtime": 60.2996, | |
| "eval_samples_per_second": 3.731, | |
| "eval_steps_per_second": 0.945, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.0204286575317383, | |
| "learning_rate": 4.569640062597809e-05, | |
| "loss": 0.7793, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 0.7613245844841003, | |
| "eval_runtime": 60.3289, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.257814407348633, | |
| "learning_rate": 4.428794992175274e-05, | |
| "loss": 0.7348, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.7604629397392273, | |
| "eval_runtime": 60.3175, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 2.0063090324401855, | |
| "learning_rate": 4.287949921752739e-05, | |
| "loss": 0.6786, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 0.7632331252098083, | |
| "eval_runtime": 60.3172, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 2.430443048477173, | |
| "learning_rate": 4.1471048513302035e-05, | |
| "loss": 0.6789, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 0.7604073882102966, | |
| "eval_runtime": 60.3176, | |
| "eval_samples_per_second": 3.73, | |
| "eval_steps_per_second": 0.945, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 1.8232342004776, | |
| "learning_rate": 4.0062597809076686e-05, | |
| "loss": 0.7177, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_loss": 0.7575519680976868, | |
| "eval_runtime": 60.3302, | |
| "eval_samples_per_second": 3.729, | |
| "eval_steps_per_second": 0.945, | |
| "step": 594 | |
| } | |
| ], | |
| "logging_steps": 54, | |
| "max_steps": 2130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 54, | |
| "total_flos": 4.802533196370739e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |