| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0024489296136813534, | |
| "eval_steps": 5, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016326197424542356, | |
| "grad_norm": 0.6254855394363403, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7351, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016326197424542356, | |
| "eval_loss": 1.7825391292572021, | |
| "eval_runtime": 350.7469, | |
| "eval_samples_per_second": 29.412, | |
| "eval_steps_per_second": 14.706, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003265239484908471, | |
| "grad_norm": 0.6790631413459778, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8095, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0004897859227362707, | |
| "grad_norm": 0.5952022075653076, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6803, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006530478969816942, | |
| "grad_norm": 0.677971363067627, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9548, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0008163098712271178, | |
| "grad_norm": 0.5969003438949585, | |
| "learning_rate": 5e-05, | |
| "loss": 1.418, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008163098712271178, | |
| "eval_loss": 1.7585334777832031, | |
| "eval_runtime": 287.1049, | |
| "eval_samples_per_second": 35.931, | |
| "eval_steps_per_second": 17.966, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0009795718454725414, | |
| "grad_norm": 0.6082297563552856, | |
| "learning_rate": 6e-05, | |
| "loss": 1.7165, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001142833819717965, | |
| "grad_norm": 0.6354915499687195, | |
| "learning_rate": 7e-05, | |
| "loss": 1.768, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0013060957939633885, | |
| "grad_norm": 0.7055122256278992, | |
| "learning_rate": 8e-05, | |
| "loss": 1.7597, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.001469357768208812, | |
| "grad_norm": 0.685647189617157, | |
| "learning_rate": 9e-05, | |
| "loss": 1.5083, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016326197424542355, | |
| "grad_norm": 0.616553008556366, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6264, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0016326197424542355, | |
| "eval_loss": 1.694636583328247, | |
| "eval_runtime": 464.5717, | |
| "eval_samples_per_second": 22.205, | |
| "eval_steps_per_second": 11.103, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0017958817166996593, | |
| "grad_norm": 0.7647472023963928, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.7135, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.001959143690945083, | |
| "grad_norm": 0.7802902460098267, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.5259, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0021224056651905064, | |
| "grad_norm": 0.7555536031723022, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.7871, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00228566763943593, | |
| "grad_norm": 0.8001874685287476, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.5847, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0024489296136813534, | |
| "grad_norm": 0.6708170771598816, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5489, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0024489296136813534, | |
| "eval_loss": 1.675947666168213, | |
| "eval_runtime": 504.0522, | |
| "eval_samples_per_second": 20.466, | |
| "eval_steps_per_second": 10.233, | |
| "step": 15 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.227876234952704e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |