| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "videomae-base-finetuned-Risky-situations/checkpoint-50", | |
| "epoch": 4.2, | |
| "eval_steps": 500, | |
| "global_step": 125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 8.420906066894531, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.8605, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 16.786991119384766, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.6079, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.9166666666666666, | |
| "eval_loss": 0.3069283962249756, | |
| "eval_runtime": 13.6517, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.44, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 8.974565505981445, | |
| "learning_rate": 4.2410714285714285e-05, | |
| "loss": 0.5951, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 9.046051979064941, | |
| "learning_rate": 3.794642857142857e-05, | |
| "loss": 0.3517, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.05629388988018036, | |
| "learning_rate": 3.348214285714286e-05, | |
| "loss": 0.1006, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00287891854532063, | |
| "eval_runtime": 13.4227, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.447, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.01983616128563881, | |
| "learning_rate": 2.9017857142857146e-05, | |
| "loss": 0.261, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.09694314002990723, | |
| "learning_rate": 2.455357142857143e-05, | |
| "loss": 0.2857, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.6325932741165161, | |
| "eval_runtime": 13.4779, | |
| "eval_samples_per_second": 0.89, | |
| "eval_steps_per_second": 0.445, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 159.1613006591797, | |
| "learning_rate": 2.0089285714285717e-05, | |
| "loss": 0.6469, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.024221086874604225, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 0.0029, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.013663964346051216, | |
| "learning_rate": 1.1160714285714287e-05, | |
| "loss": 0.0044, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0005676120636053383, | |
| "eval_runtime": 13.7149, | |
| "eval_samples_per_second": 0.875, | |
| "eval_steps_per_second": 0.437, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 385.0979309082031, | |
| "learning_rate": 6.696428571428572e-06, | |
| "loss": 0.1138, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 0.011926279403269291, | |
| "learning_rate": 2.2321428571428573e-06, | |
| "loss": 0.0007, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00049240468069911, | |
| "eval_runtime": 13.8419, | |
| "eval_samples_per_second": 0.867, | |
| "eval_steps_per_second": 0.433, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "step": 125, | |
| "total_flos": 3.140082765468795e+17, | |
| "train_loss": 0.3065260285772383, | |
| "train_runtime": 512.4464, | |
| "train_samples_per_second": 0.488, | |
| "train_steps_per_second": 0.244 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.9399034380912781, | |
| "eval_runtime": 13.7322, | |
| "eval_samples_per_second": 0.874, | |
| "eval_steps_per_second": 0.437, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.00287891854532063, | |
| "eval_runtime": 14.128, | |
| "eval_samples_per_second": 0.849, | |
| "eval_steps_per_second": 0.425, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.9399035573005676, | |
| "eval_runtime": 13.5901, | |
| "eval_samples_per_second": 0.883, | |
| "eval_steps_per_second": 0.441, | |
| "step": 125 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "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": 3.140082765468795e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |