| { | |
| "best_metric": 0.46333640813827515, | |
| "best_model_checkpoint": "realFake-img/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": 5.288677215576172, | |
| "learning_rate": 0.00019038461538461538, | |
| "loss": 0.1567, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 4.266819953918457, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 0.2467, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 3.213836193084717, | |
| "learning_rate": 0.00017115384615384616, | |
| "loss": 0.3028, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.7542410492897034, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.1968, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 11.869466781616211, | |
| "learning_rate": 0.00015192307692307692, | |
| "loss": 0.2678, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 0.7070229649543762, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 0.1494, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 3.979475975036621, | |
| "learning_rate": 0.0001326923076923077, | |
| "loss": 0.0927, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 6.856090545654297, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.2443, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 0.7266550064086914, | |
| "learning_rate": 0.00011346153846153846, | |
| "loss": 0.1968, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 3.321963310241699, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 0.1137, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "eval_accuracy": 0.8287671232876712, | |
| "eval_loss": 0.48689645528793335, | |
| "eval_runtime": 5.0712, | |
| "eval_samples_per_second": 28.79, | |
| "eval_steps_per_second": 3.747, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 6.398293972015381, | |
| "learning_rate": 9.423076923076924e-05, | |
| "loss": 0.0703, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.38959336280822754, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.1407, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.1938396394252777, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1098, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 5.93131685256958, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 0.0969, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 0.0541561059653759, | |
| "learning_rate": 5.576923076923077e-05, | |
| "loss": 0.1299, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 3.552217960357666, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.1264, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 8.6173095703125, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.1076, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 0.7253573536872864, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.157, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 7.486116409301758, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.177, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.391594409942627, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1002, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_accuracy": 0.8835616438356164, | |
| "eval_loss": 0.46333640813827515, | |
| "eval_runtime": 4.8332, | |
| "eval_samples_per_second": 30.208, | |
| "eval_steps_per_second": 3.931, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 208, | |
| "total_flos": 2.5832369176982323e+17, | |
| "train_loss": 0.1598776997281955, | |
| "train_runtime": 141.5375, | |
| "train_samples_per_second": 23.344, | |
| "train_steps_per_second": 1.47 | |
| } | |
| ], | |
| "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": 2.5832369176982323e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |