| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.914285714285714, | |
| "eval_steps": 500, | |
| "global_step": 51, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 7.817991325483718, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.965, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 49.73347603245408, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.1198, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 4.523926056457663, | |
| "learning_rate": 1.961261695938319e-05, | |
| "loss": 1.0209, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 3.8093568875972696, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 1.0022, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 1.349517822265625, | |
| "eval_runtime": 2.9563, | |
| "eval_samples_per_second": 7.78, | |
| "eval_steps_per_second": 0.677, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 2.9739767130682138, | |
| "learning_rate": 1.5591929034707468e-05, | |
| "loss": 0.8913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 2.5183506153336306, | |
| "learning_rate": 1.2419218955996677e-05, | |
| "loss": 0.8043, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 2.193683108302545, | |
| "learning_rate": 8.954715367323468e-06, | |
| "loss": 0.7194, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.3903108778369, | |
| "learning_rate": 5.616288532109225e-06, | |
| "loss": 0.7519, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3249517679214478, | |
| "eval_runtime": 2.7504, | |
| "eval_samples_per_second": 8.363, | |
| "eval_steps_per_second": 0.727, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 3.1793272954159764, | |
| "learning_rate": 2.8066019966134907e-06, | |
| "loss": 0.6556, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 3.802719468935368, | |
| "learning_rate": 8.645454235739903e-07, | |
| "loss": 0.6099, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 2.3574393295768283, | |
| "learning_rate": 2.4359497401758026e-08, | |
| "loss": 0.5685, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 1.3446100950241089, | |
| "eval_runtime": 2.6623, | |
| "eval_samples_per_second": 8.639, | |
| "eval_steps_per_second": 0.751, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "step": 51, | |
| "total_flos": 10626017525760.0, | |
| "train_loss": 0.807617746147455, | |
| "train_runtime": 901.1485, | |
| "train_samples_per_second": 3.685, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 51, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 10626017525760.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |