| { | |
| "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": 189.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 47.9246, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_loss": 15.484689712524414, | |
| "eval_runtime": 0.2675, | |
| "eval_samples_per_second": 37.382, | |
| "eval_steps_per_second": 3.738, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 52.25, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 38.5919, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 12.549888610839844, | |
| "eval_runtime": 0.2336, | |
| "eval_samples_per_second": 42.806, | |
| "eval_steps_per_second": 4.281, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 11.375, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 23.0632, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_loss": 10.153787612915039, | |
| "eval_runtime": 0.2414, | |
| "eval_samples_per_second": 41.423, | |
| "eval_steps_per_second": 4.142, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 8.464659690856934, | |
| "eval_runtime": 0.2419, | |
| "eval_samples_per_second": 41.336, | |
| "eval_steps_per_second": 4.134, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 19.8584, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_loss": 8.02161979675293, | |
| "eval_runtime": 0.2493, | |
| "eval_samples_per_second": 40.12, | |
| "eval_steps_per_second": 4.012, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 19.1062, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 7.756901741027832, | |
| "eval_runtime": 0.2345, | |
| "eval_samples_per_second": 42.648, | |
| "eval_steps_per_second": 4.265, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_loss": 7.677903175354004, | |
| "eval_runtime": 0.25, | |
| "eval_samples_per_second": 40.001, | |
| "eval_steps_per_second": 4.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 4.46875, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 18.5688, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 7.643762111663818, | |
| "eval_runtime": 0.2433, | |
| "eval_samples_per_second": 41.101, | |
| "eval_steps_per_second": 4.11, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 4.84375, | |
| "learning_rate": 0.0, | |
| "loss": 18.5805, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_loss": 7.6471662521362305, | |
| "eval_runtime": 0.2333, | |
| "eval_samples_per_second": 42.859, | |
| "eval_steps_per_second": 4.286, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "step": 30, | |
| "total_flos": 9.147533553041408e+16, | |
| "train_loss": 23.272574106852215, | |
| "train_runtime": 74.8732, | |
| "train_samples_per_second": 29.383, | |
| "train_steps_per_second": 0.401 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 9.147533553041408e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |