| { | |
| "best_metric": 5.2793288230896, | |
| "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained/de_clm/childes_30/checkpoint-8000", | |
| "epoch": 8.385744234800839, | |
| "eval_steps": 2000, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.0964360587002098, | |
| "eval_loss": 7.102903366088867, | |
| "eval_runtime": 0.9708, | |
| "eval_samples_per_second": 1416.286, | |
| "eval_steps_per_second": 88.582, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.1928721174004195, | |
| "grad_norm": 1.3964662551879883, | |
| "learning_rate": 1e-05, | |
| "loss": 6.9987, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.1928721174004195, | |
| "eval_loss": 5.884151935577393, | |
| "eval_runtime": 0.966, | |
| "eval_samples_per_second": 1423.408, | |
| "eval_steps_per_second": 89.028, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "eval_loss": 5.54873514175415, | |
| "eval_runtime": 0.9657, | |
| "eval_samples_per_second": 1423.84, | |
| "eval_steps_per_second": 89.055, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.385744234800839, | |
| "grad_norm": 2.7172107696533203, | |
| "learning_rate": 2e-05, | |
| "loss": 5.2204, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.385744234800839, | |
| "eval_loss": 5.2793288230896, | |
| "eval_runtime": 0.9644, | |
| "eval_samples_per_second": 1425.779, | |
| "eval_steps_per_second": 89.176, | |
| "step": 8000 | |
| } | |
| ], | |
| "logging_steps": 4000, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 105, | |
| "save_steps": 4000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2068529059479552.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |