| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9863013698630136, | |
| "eval_steps": 2, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0, | |
| "loss": 1.5513, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 1.6200196743011475, | |
| "eval_runtime": 6.6812, | |
| "eval_samples_per_second": 37.418, | |
| "eval_steps_per_second": 2.395, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5555, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 1.6200196743011475, | |
| "eval_runtime": 6.6725, | |
| "eval_samples_per_second": 37.467, | |
| "eval_steps_per_second": 2.398, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4e-05, | |
| "loss": 1.559, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 6e-05, | |
| "loss": 1.5558, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.618035078048706, | |
| "eval_runtime": 6.6887, | |
| "eval_samples_per_second": 37.377, | |
| "eval_steps_per_second": 2.392, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8e-05, | |
| "loss": 1.5365, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5195, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.6109036207199097, | |
| "eval_runtime": 6.677, | |
| "eval_samples_per_second": 37.442, | |
| "eval_steps_per_second": 2.396, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00012, | |
| "loss": 1.5266, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00014, | |
| "loss": 1.5358, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.592940092086792, | |
| "eval_runtime": 6.6867, | |
| "eval_samples_per_second": 37.388, | |
| "eval_steps_per_second": 2.393, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00016, | |
| "loss": 1.4989, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00018, | |
| "loss": 1.5124, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 1.573974609375, | |
| "eval_runtime": 6.6699, | |
| "eval_samples_per_second": 37.482, | |
| "eval_steps_per_second": 2.399, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.502, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.4938, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.5591483116149902, | |
| "eval_runtime": 6.6798, | |
| "eval_samples_per_second": 37.427, | |
| "eval_steps_per_second": 2.395, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 1.5116, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 1.4881, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_loss": 1.5495102405548096, | |
| "eval_runtime": 6.6802, | |
| "eval_samples_per_second": 37.424, | |
| "eval_steps_per_second": 2.395, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 1.5033, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 1.4639, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 1.542672872543335, | |
| "eval_runtime": 6.6782, | |
| "eval_samples_per_second": 37.435, | |
| "eval_steps_per_second": 2.396, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.465, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 1.4824, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 1.537250280380249, | |
| "eval_runtime": 6.677, | |
| "eval_samples_per_second": 37.442, | |
| "eval_steps_per_second": 2.396, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 36, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 1.4900298798973583e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |