| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00218835140546869, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.3767028109373806e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1118.6366, | |
| "eval_samples_per_second": 8.601, | |
| "eval_steps_per_second": 4.301, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00021883514054686903, | |
| "grad_norm": 149.75144958496094, | |
| "learning_rate": 5e-05, | |
| "loss": 44.7515, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00043767028109373805, | |
| "grad_norm": 97.91552734375, | |
| "learning_rate": 0.0001, | |
| "loss": 33.6816, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00043767028109373805, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1117.0239, | |
| "eval_samples_per_second": 8.613, | |
| "eval_steps_per_second": 4.307, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000656505421640607, | |
| "grad_norm": 76.67736053466797, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 18.14, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0008753405621874761, | |
| "grad_norm": 61.07263946533203, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 11.8059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008753405621874761, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1116.9665, | |
| "eval_samples_per_second": 8.614, | |
| "eval_steps_per_second": 4.307, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001094175702734345, | |
| "grad_norm": 32.5378532409668, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 9.8131, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001313010843281214, | |
| "grad_norm": 59.290103912353516, | |
| "learning_rate": 5e-05, | |
| "loss": 8.5577, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001313010843281214, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1116.704, | |
| "eval_samples_per_second": 8.616, | |
| "eval_steps_per_second": 4.308, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001531845983828083, | |
| "grad_norm": 94.53697967529297, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 5.2927, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0017506811243749522, | |
| "grad_norm": 63.27540969848633, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 3.9622, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0017506811243749522, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1117.7667, | |
| "eval_samples_per_second": 8.607, | |
| "eval_steps_per_second": 4.304, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001969516264921821, | |
| "grad_norm": 97.67462158203125, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 3.3854, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00218835140546869, | |
| "grad_norm": 109.79862976074219, | |
| "learning_rate": 0.0, | |
| "loss": 2.933, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00218835140546869, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1116.4721, | |
| "eval_samples_per_second": 8.617, | |
| "eval_steps_per_second": 4.309, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7482881114832896e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |