| { | |
| "best_metric": 0.35287705063819885, | |
| "best_model_checkpoint": "realFake-food/checkpoint-200", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 208, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 1.8828963041305542, | |
| "learning_rate": 0.00019038461538461538, | |
| "loss": 0.998, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 6.118525505065918, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 0.5874, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 3.4514338970184326, | |
| "learning_rate": 0.00017115384615384616, | |
| "loss": 0.6629, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.1951334476470947, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.5009, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 3.5224995613098145, | |
| "learning_rate": 0.00015192307692307692, | |
| "loss": 0.501, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.2696192264556885, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 0.3405, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 4.657229900360107, | |
| "learning_rate": 0.0001326923076923077, | |
| "loss": 0.4532, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.9673850536346436, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.4317, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 2.4534852504730225, | |
| "learning_rate": 0.00011346153846153846, | |
| "loss": 0.3651, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 2.316755771636963, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 0.432, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "eval_accuracy": 0.821917808219178, | |
| "eval_loss": 0.3703341484069824, | |
| "eval_runtime": 4.378, | |
| "eval_samples_per_second": 33.349, | |
| "eval_steps_per_second": 4.34, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 1.4332585334777832, | |
| "learning_rate": 9.423076923076924e-05, | |
| "loss": 0.2974, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 2.3531041145324707, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.2768, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.171938180923462, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.2914, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 3.204784631729126, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 0.3011, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 5.116678714752197, | |
| "learning_rate": 5.576923076923077e-05, | |
| "loss": 0.2201, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.1804983615875244, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.341, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 0.4275173842906952, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.1692, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 2.864809036254883, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.1823, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 1.2293394804000854, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.1454, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 3.070647954940796, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1588, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_accuracy": 0.8561643835616438, | |
| "eval_loss": 0.35287705063819885, | |
| "eval_runtime": 5.6811, | |
| "eval_samples_per_second": 25.699, | |
| "eval_steps_per_second": 3.344, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 208, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_loss": 0.3738819756186925, | |
| "train_runtime": 179.0456, | |
| "train_samples_per_second": 18.453, | |
| "train_steps_per_second": 1.162 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 208, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |