| { | |
| "best_metric": 0.9622508792497069, | |
| "best_model_checkpoint": "ky-finetuned-skindiseaseicthuawei32/checkpoint-3000", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 135.11822509765625, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3894, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8060961313012895, | |
| "eval_loss": 0.6159797310829163, | |
| "eval_runtime": 59.1496, | |
| "eval_samples_per_second": 72.105, | |
| "eval_steps_per_second": 2.265, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 114.53544616699219, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.6543, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8635404454865182, | |
| "eval_loss": 0.43777740001678467, | |
| "eval_runtime": 58.9516, | |
| "eval_samples_per_second": 72.348, | |
| "eval_steps_per_second": 2.273, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 72.00625610351562, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.471, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9160609613130128, | |
| "eval_loss": 0.2565539479255676, | |
| "eval_runtime": 59.0158, | |
| "eval_samples_per_second": 72.269, | |
| "eval_steps_per_second": 2.271, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 76.05452728271484, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3853, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9134818288393904, | |
| "eval_loss": 0.2498469352722168, | |
| "eval_runtime": 59.1502, | |
| "eval_samples_per_second": 72.105, | |
| "eval_steps_per_second": 2.265, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 110.07782745361328, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3225, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9289566236811254, | |
| "eval_loss": 0.21574878692626953, | |
| "eval_runtime": 59.1951, | |
| "eval_samples_per_second": 72.05, | |
| "eval_steps_per_second": 2.264, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 54.6344108581543, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2769, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9406799531066823, | |
| "eval_loss": 0.1746564507484436, | |
| "eval_runtime": 59.244, | |
| "eval_samples_per_second": 71.99, | |
| "eval_steps_per_second": 2.262, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 49.57483673095703, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.2364, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.948651817116061, | |
| "eval_loss": 0.15020643174648285, | |
| "eval_runtime": 59.6495, | |
| "eval_samples_per_second": 71.501, | |
| "eval_steps_per_second": 2.246, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 85.14453125, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.2005, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9547479484173506, | |
| "eval_loss": 0.12820766866207123, | |
| "eval_runtime": 59.7687, | |
| "eval_samples_per_second": 71.358, | |
| "eval_steps_per_second": 2.242, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 20.916675567626953, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1737, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9596717467760844, | |
| "eval_loss": 0.11289093643426895, | |
| "eval_runtime": 59.3074, | |
| "eval_samples_per_second": 71.913, | |
| "eval_steps_per_second": 2.259, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 45.390342712402344, | |
| "learning_rate": 0.0, | |
| "loss": 0.1468, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9622508792497069, | |
| "eval_loss": 0.10576048493385315, | |
| "eval_runtime": 59.8846, | |
| "eval_samples_per_second": 71.22, | |
| "eval_steps_per_second": 2.238, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 3000, | |
| "total_flos": 3.922659491677471e+19, | |
| "train_loss": 0.4256939633687337, | |
| "train_runtime": 8577.8057, | |
| "train_samples_per_second": 44.749, | |
| "train_steps_per_second": 0.35 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 3.922659491677471e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |