| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12371134020618557, | |
| "eval_steps": 9, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003436426116838488, | |
| "eval_loss": 1.0753878355026245, | |
| "eval_runtime": 70.8712, | |
| "eval_samples_per_second": 6.914, | |
| "eval_steps_per_second": 0.875, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010309278350515464, | |
| "grad_norm": 9.508790016174316, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2728, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 4.23081111907959, | |
| "learning_rate": 6e-05, | |
| "loss": 4.2308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "grad_norm": 2.994974136352539, | |
| "learning_rate": 9e-05, | |
| "loss": 4.0505, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "eval_loss": 0.9270894527435303, | |
| "eval_runtime": 72.0129, | |
| "eval_samples_per_second": 6.804, | |
| "eval_steps_per_second": 0.861, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041237113402061855, | |
| "grad_norm": 3.4861974716186523, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.3622, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05154639175257732, | |
| "grad_norm": 2.4739320278167725, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.4222, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "grad_norm": 2.656764507293701, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.3192, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "eval_loss": 0.781949520111084, | |
| "eval_runtime": 72.0068, | |
| "eval_samples_per_second": 6.805, | |
| "eval_steps_per_second": 0.861, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07216494845360824, | |
| "grad_norm": 2.133312702178955, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.9592, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08247422680412371, | |
| "grad_norm": 2.0654103755950928, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.2154, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "grad_norm": 1.8565651178359985, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.9275, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "eval_loss": 0.7440634965896606, | |
| "eval_runtime": 72.0175, | |
| "eval_samples_per_second": 6.804, | |
| "eval_steps_per_second": 0.861, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 2.0615859031677246, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.9618, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1134020618556701, | |
| "grad_norm": 2.0000863075256348, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.0231, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 1.8480544090270996, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.8036, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "eval_loss": 0.727074146270752, | |
| "eval_runtime": 72.0101, | |
| "eval_samples_per_second": 6.805, | |
| "eval_steps_per_second": 0.861, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2808770639888384e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |