| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.021691973969631236, | |
| "eval_steps": 4, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0010845986984815619, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010845986984815619, | |
| "eval_loss": NaN, | |
| "eval_runtime": 51.7259, | |
| "eval_samples_per_second": 7.52, | |
| "eval_steps_per_second": 3.77, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0021691973969631237, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0032537960954446853, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004338394793926247, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004338394793926247, | |
| "eval_loss": NaN, | |
| "eval_runtime": 54.4666, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 3.58, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005422993492407809, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006507592190889371, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007592190889370932, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008676789587852495, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.008676789587852495, | |
| "eval_loss": NaN, | |
| "eval_runtime": 80.2469, | |
| "eval_samples_per_second": 4.848, | |
| "eval_steps_per_second": 2.43, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009761388286334056, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.010845986984815618, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01193058568329718, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013015184381778741, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.013015184381778741, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8068, | |
| "eval_samples_per_second": 7.656, | |
| "eval_steps_per_second": 3.838, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014099783080260303, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015184381778741865, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.016268980477223426, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01735357917570499, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01735357917570499, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8293, | |
| "eval_samples_per_second": 7.653, | |
| "eval_steps_per_second": 3.836, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01843817787418655, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.019522776572668113, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020607375271149676, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.021691973969631236, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021691973969631236, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8357, | |
| "eval_samples_per_second": 7.652, | |
| "eval_steps_per_second": 3.836, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5589685858467840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |