{ "best_metric": 0.9660714285714286, "best_model_checkpoint": "vit-large-patch16-224-in21k-dungeon-geo-morphs-0-4-30Nov24-002/checkpoint-70", "epoch": 31.90909090909091, "eval_steps": 10, "global_step": 80, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 3.909090909090909, "grad_norm": 14.049148559570312, "learning_rate": 9.861111111111112e-06, "loss": 1.517, "step": 10 }, { "epoch": 3.909090909090909, "eval_accuracy": 0.6767857142857143, "eval_loss": 1.3385733366012573, "eval_runtime": 7.4822, "eval_samples_per_second": 74.845, "eval_steps_per_second": 9.356, "step": 10 }, { "epoch": 7.909090909090909, "grad_norm": 12.874731063842773, "learning_rate": 8.472222222222223e-06, "loss": 0.8959, "step": 20 }, { "epoch": 7.909090909090909, "eval_accuracy": 0.9089285714285714, "eval_loss": 0.887944221496582, "eval_runtime": 7.8695, "eval_samples_per_second": 71.161, "eval_steps_per_second": 8.895, "step": 20 }, { "epoch": 11.909090909090908, "grad_norm": 7.380621433258057, "learning_rate": 7.083333333333335e-06, "loss": 0.4053, "step": 30 }, { "epoch": 11.909090909090908, "eval_accuracy": 0.9375, "eval_loss": 0.5938794612884521, "eval_runtime": 7.8248, "eval_samples_per_second": 71.567, "eval_steps_per_second": 8.946, "step": 30 }, { "epoch": 15.909090909090908, "grad_norm": 4.831039905548096, "learning_rate": 5.694444444444445e-06, "loss": 0.173, "step": 40 }, { "epoch": 15.909090909090908, "eval_accuracy": 0.95, "eval_loss": 0.4381349980831146, "eval_runtime": 7.4435, "eval_samples_per_second": 75.234, "eval_steps_per_second": 9.404, "step": 40 }, { "epoch": 19.90909090909091, "grad_norm": 2.24302339553833, "learning_rate": 4.305555555555556e-06, "loss": 0.0766, "step": 50 }, { "epoch": 19.90909090909091, "eval_accuracy": 0.9589285714285715, "eval_loss": 0.33938032388687134, "eval_runtime": 8.1587, "eval_samples_per_second": 68.638, "eval_steps_per_second": 8.58, "step": 50 }, { "epoch": 23.90909090909091, "grad_norm": 1.3549830913543701, "learning_rate": 2.916666666666667e-06, "loss": 0.0395, "step": 60 }, { "epoch": 23.90909090909091, "eval_accuracy": 0.9642857142857143, "eval_loss": 0.28536272048950195, "eval_runtime": 7.0653, "eval_samples_per_second": 79.261, "eval_steps_per_second": 9.908, "step": 60 }, { "epoch": 27.90909090909091, "grad_norm": 0.6997795104980469, "learning_rate": 1.527777777777778e-06, "loss": 0.0243, "step": 70 }, { "epoch": 27.90909090909091, "eval_accuracy": 0.9660714285714286, "eval_loss": 0.2581222951412201, "eval_runtime": 7.9504, "eval_samples_per_second": 70.437, "eval_steps_per_second": 8.805, "step": 70 }, { "epoch": 31.90909090909091, "grad_norm": 0.569978654384613, "learning_rate": 1.3888888888888888e-07, "loss": 0.0186, "step": 80 }, { "epoch": 31.90909090909091, "eval_accuracy": 0.9660714285714286, "eval_loss": 0.24860888719558716, "eval_runtime": 8.1493, "eval_samples_per_second": 68.717, "eval_steps_per_second": 8.59, "step": 80 }, { "epoch": 31.90909090909091, "step": 80, "total_flos": 7.012786101918106e+17, "train_loss": 0.39377664029598236, "train_runtime": 678.5627, "train_samples_per_second": 4.834, "train_steps_per_second": 0.118 } ], "logging_steps": 10, "max_steps": 80, "num_input_tokens_seen": 0, "num_train_epochs": 40, "save_steps": 10, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.012786101918106e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }