| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 20, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "eval_loss": 2.15944766998291, | |
| "eval_runtime": 15.0273, | |
| "eval_samples_per_second": 99.885, | |
| "eval_steps_per_second": 5.057, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 1.9376, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.93359375, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.835, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.723530888557434, | |
| "eval_runtime": 14.1976, | |
| "eval_samples_per_second": 105.722, | |
| "eval_steps_per_second": 5.353, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9109, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 0.00019984815164333163, | |
| "loss": 1.9526, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.8265540599822998, | |
| "eval_runtime": 15.0951, | |
| "eval_samples_per_second": 99.436, | |
| "eval_steps_per_second": 5.035, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.0546875, | |
| "learning_rate": 0.00019939306773179497, | |
| "loss": 1.1096, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.0, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 1.0699, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 2.048560619354248, | |
| "eval_runtime": 14.1768, | |
| "eval_samples_per_second": 105.877, | |
| "eval_steps_per_second": 5.361, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.984375, | |
| "learning_rate": 0.00019757963826274357, | |
| "loss": 1.0686, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00019622680003092503, | |
| "loss": 1.1259, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.0676190853118896, | |
| "eval_runtime": 14.1344, | |
| "eval_samples_per_second": 106.194, | |
| "eval_steps_per_second": 5.377, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 0.4202, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 0.00019264940672148018, | |
| "loss": 0.4238, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 2.4824252128601074, | |
| "eval_runtime": 14.1808, | |
| "eval_samples_per_second": 105.847, | |
| "eval_steps_per_second": 5.359, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 0.00019043571606975777, | |
| "loss": 0.4378, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 0.4412, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.4695043563842773, | |
| "eval_runtime": 14.1704, | |
| "eval_samples_per_second": 105.925, | |
| "eval_steps_per_second": 5.363, | |
| "step": 120 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.774499349764506e+16, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |