| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0028238217603704856, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.647643520740971e-05, | |
| "eval_loss": 18.57196617126465, | |
| "eval_runtime": 916.5457, | |
| "eval_samples_per_second": 8.135, | |
| "eval_steps_per_second": 4.067, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00028238217603704857, | |
| "grad_norm": 7.614293575286865, | |
| "learning_rate": 5e-05, | |
| "loss": 18.4224, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005647643520740971, | |
| "grad_norm": 17.468368530273438, | |
| "learning_rate": 0.0001, | |
| "loss": 17.7677, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005647643520740971, | |
| "eval_loss": 15.393836975097656, | |
| "eval_runtime": 920.5368, | |
| "eval_samples_per_second": 8.1, | |
| "eval_steps_per_second": 4.05, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0008471465281111456, | |
| "grad_norm": 22.757177352905273, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 11.6793, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0011295287041481943, | |
| "grad_norm": 14.26730728149414, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.5892, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0011295287041481943, | |
| "eval_loss": 1.6451387405395508, | |
| "eval_runtime": 919.4827, | |
| "eval_samples_per_second": 8.109, | |
| "eval_steps_per_second": 4.054, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014119108801852428, | |
| "grad_norm": 7.687915802001953, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.1298, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0016942930562222913, | |
| "grad_norm": 6.556017875671387, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1151, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0016942930562222913, | |
| "eval_loss": 1.0435093641281128, | |
| "eval_runtime": 920.9656, | |
| "eval_samples_per_second": 8.096, | |
| "eval_steps_per_second": 4.048, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0019766752322593396, | |
| "grad_norm": 6.714224815368652, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.9979, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0022590574082963885, | |
| "grad_norm": 6.470224380493164, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.8955, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0022590574082963885, | |
| "eval_loss": 1.0277140140533447, | |
| "eval_runtime": 924.6909, | |
| "eval_samples_per_second": 8.063, | |
| "eval_steps_per_second": 4.032, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.002541439584333437, | |
| "grad_norm": 6.631415367126465, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.1846, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0028238217603704856, | |
| "grad_norm": 4.103190898895264, | |
| "learning_rate": 0.0, | |
| "loss": 0.9464, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0028238217603704856, | |
| "eval_loss": 1.0135915279388428, | |
| "eval_runtime": 921.0176, | |
| "eval_samples_per_second": 8.095, | |
| "eval_steps_per_second": 4.048, | |
| "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": 9247821240729600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |