| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.571428571428571, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.9846, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_loss": 3.1614296436309814, | |
| "eval_runtime": 0.5389, | |
| "eval_samples_per_second": 18.556, | |
| "eval_steps_per_second": 1.856, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 2.8305, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.8875155448913574, | |
| "eval_runtime": 0.51, | |
| "eval_samples_per_second": 19.607, | |
| "eval_steps_per_second": 1.961, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 2.3403, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_loss": 2.834296703338623, | |
| "eval_runtime": 0.626, | |
| "eval_samples_per_second": 15.975, | |
| "eval_steps_per_second": 1.597, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.7485246658325195, | |
| "eval_runtime": 0.5273, | |
| "eval_samples_per_second": 18.965, | |
| "eval_steps_per_second": 1.896, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 2.0675, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_loss": 2.6912894248962402, | |
| "eval_runtime": 0.6579, | |
| "eval_samples_per_second": 15.199, | |
| "eval_steps_per_second": 1.52, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 1.8758, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.6605114936828613, | |
| "eval_runtime": 0.5348, | |
| "eval_samples_per_second": 18.698, | |
| "eval_steps_per_second": 1.87, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_loss": 2.648810625076294, | |
| "eval_runtime": 0.6999, | |
| "eval_samples_per_second": 14.288, | |
| "eval_steps_per_second": 1.429, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 2.375, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 1.7659, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.645064353942871, | |
| "eval_runtime": 0.543, | |
| "eval_samples_per_second": 18.416, | |
| "eval_steps_per_second": 1.842, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0, | |
| "loss": 1.7298, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_loss": 2.6449332237243652, | |
| "eval_runtime": 0.5444, | |
| "eval_samples_per_second": 18.369, | |
| "eval_steps_per_second": 1.837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "step": 30, | |
| "total_flos": 2.34949340823552e+16, | |
| "train_loss": 2.1067513783772784, | |
| "train_runtime": 107.4922, | |
| "train_samples_per_second": 20.281, | |
| "train_steps_per_second": 0.279 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 2.34949340823552e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |