| { |
| "best_global_step": 98, |
| "best_metric": 0.8295379281044006, |
| "best_model_checkpoint": "/workspace/scripts/soutputs/fa30d435-3229-4c56-871d-51ad38cfbe7a_7/checkpoint-98", |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 98, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.10204081632653061, |
| "grad_norm": 0.74609375, |
| "learning_rate": 0.0005517817935364519, |
| "loss": 1.267, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.20408163265306123, |
| "grad_norm": 0.43359375, |
| "learning_rate": 0.0006868981122853327, |
| "loss": 1.133, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.30612244897959184, |
| "grad_norm": 0.3515625, |
| "learning_rate": 0.0006754931100325356, |
| "loss": 1.0545, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.40816326530612246, |
| "grad_norm": 0.29296875, |
| "learning_rate": 0.0006556597531537671, |
| "loss": 0.9658, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.5102040816326531, |
| "grad_norm": 0.25390625, |
| "learning_rate": 0.0006279625071049859, |
| "loss": 0.9251, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.6122448979591837, |
| "grad_norm": 0.23046875, |
| "learning_rate": 0.0005931896468471727, |
| "loss": 0.9068, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.7142857142857143, |
| "grad_norm": 0.2734375, |
| "learning_rate": 0.0005523308222182446, |
| "loss": 0.8653, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.8163265306122449, |
| "grad_norm": 0.259765625, |
| "learning_rate": 0.0005065488920935338, |
| "loss": 0.8615, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.9183673469387755, |
| "grad_norm": 0.2255859375, |
| "learning_rate": 0.00045714682894615636, |
| "loss": 0.8336, |
| "step": 45 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.8490258455276489, |
| "eval_runtime": 7.2501, |
| "eval_samples_per_second": 27.448, |
| "eval_steps_per_second": 27.448, |
| "step": 49 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.8490258455276489, |
| "eval_runtime": 7.2144, |
| "eval_samples_per_second": 27.584, |
| "eval_steps_per_second": 27.584, |
| "step": 49 |
| }, |
| { |
| "epoch": 1.0204081632653061, |
| "grad_norm": 0.51171875, |
| "learning_rate": 0.0004055306357170286, |
| "loss": 0.7853, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.1224489795918366, |
| "grad_norm": 0.271484375, |
| "learning_rate": 0.00035316933039559285, |
| "loss": 0.5521, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.2244897959183674, |
| "grad_norm": 0.265625, |
| "learning_rate": 0.00030155313716646496, |
| "loss": 0.5756, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.3265306122448979, |
| "grad_norm": 0.26953125, |
| "learning_rate": 0.00025215107401908756, |
| "loss": 0.5582, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.4285714285714286, |
| "grad_norm": 0.224609375, |
| "learning_rate": 0.00020636914389437675, |
| "loss": 0.5496, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.5306122448979593, |
| "grad_norm": 0.2275390625, |
| "learning_rate": 0.0001655103192654486, |
| "loss": 0.5427, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.6326530612244898, |
| "grad_norm": 0.267578125, |
| "learning_rate": 0.00013073745900763563, |
| "loss": 0.5402, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.7346938775510203, |
| "grad_norm": 0.19140625, |
| "learning_rate": 0.00010304021295885433, |
| "loss": 0.5118, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.836734693877551, |
| "grad_norm": 0.251953125, |
| "learning_rate": 8.320685608008573e-05, |
| "loss": 0.5408, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.9387755102040818, |
| "grad_norm": 0.234375, |
| "learning_rate": 7.18018538272886e-05, |
| "loss": 0.5387, |
| "step": 95 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.8295379281044006, |
| "eval_runtime": 7.3772, |
| "eval_samples_per_second": 26.975, |
| "eval_steps_per_second": 26.975, |
| "step": 98 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 98, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.939659357683712e+17, |
| "train_batch_size": 100, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
