| { | |
| "best_metric": 0.9857142857142858, | |
| "best_model_checkpoint": "document-spoof-clip/checkpoint-9", | |
| "epoch": 16.842105263157894, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8421052631578947, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.2112182229757309, | |
| "eval_runtime": 5.2092, | |
| "eval_samples_per_second": 13.438, | |
| "eval_steps_per_second": 0.96, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "eval_accuracy": 0.9857142857142858, | |
| "eval_loss": 0.12266874313354492, | |
| "eval_runtime": 5.2038, | |
| "eval_samples_per_second": 13.452, | |
| "eval_steps_per_second": 0.961, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 4.123354434967041, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.295, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.9473684210526314, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.12026520818471909, | |
| "eval_runtime": 5.3187, | |
| "eval_samples_per_second": 13.161, | |
| "eval_steps_per_second": 0.94, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9714285714285714, | |
| "eval_loss": 0.06349408626556396, | |
| "eval_runtime": 5.1916, | |
| "eval_samples_per_second": 13.483, | |
| "eval_steps_per_second": 0.963, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 14.247283935546875, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0962, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.842105263157895, | |
| "eval_accuracy": 0.9428571428571428, | |
| "eval_loss": 0.2938525080680847, | |
| "eval_runtime": 5.3233, | |
| "eval_samples_per_second": 13.15, | |
| "eval_steps_per_second": 0.939, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 5.894736842105263, | |
| "eval_accuracy": 0.9285714285714286, | |
| "eval_loss": 0.24831673502922058, | |
| "eval_runtime": 5.2672, | |
| "eval_samples_per_second": 13.29, | |
| "eval_steps_per_second": 0.949, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 36.67098617553711, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.163, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.947368421052632, | |
| "eval_accuracy": 0.9857142857142858, | |
| "eval_loss": 0.07119516283273697, | |
| "eval_runtime": 5.2492, | |
| "eval_samples_per_second": 13.335, | |
| "eval_steps_per_second": 0.953, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9714285714285714, | |
| "eval_loss": 0.04743731766939163, | |
| "eval_runtime": 5.2254, | |
| "eval_samples_per_second": 13.396, | |
| "eval_steps_per_second": 0.957, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 14.154749870300293, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.842105263157894, | |
| "eval_accuracy": 0.9428571428571428, | |
| "eval_loss": 0.20123328268527985, | |
| "eval_runtime": 5.2671, | |
| "eval_samples_per_second": 13.29, | |
| "eval_steps_per_second": 0.949, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 9.894736842105264, | |
| "eval_accuracy": 0.9, | |
| "eval_loss": 0.3587024211883545, | |
| "eval_runtime": 5.2483, | |
| "eval_samples_per_second": 13.338, | |
| "eval_steps_per_second": 0.953, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 10.526315789473685, | |
| "grad_norm": 3.9370455741882324, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1048, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.947368421052632, | |
| "eval_accuracy": 0.9857142857142858, | |
| "eval_loss": 0.04267547279596329, | |
| "eval_runtime": 5.2718, | |
| "eval_samples_per_second": 13.278, | |
| "eval_steps_per_second": 0.948, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9857142857142858, | |
| "eval_loss": 0.01490915659815073, | |
| "eval_runtime": 5.2034, | |
| "eval_samples_per_second": 13.453, | |
| "eval_steps_per_second": 0.961, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 12.631578947368421, | |
| "grad_norm": 5.009355545043945, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0519, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 12.842105263157894, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.16158795356750488, | |
| "eval_runtime": 5.284, | |
| "eval_samples_per_second": 13.248, | |
| "eval_steps_per_second": 0.946, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 13.894736842105264, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.2285740226507187, | |
| "eval_runtime": 5.3185, | |
| "eval_samples_per_second": 13.161, | |
| "eval_steps_per_second": 0.94, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 14.736842105263158, | |
| "grad_norm": 1.6222240924835205, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0151, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 14.947368421052632, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.13689564168453217, | |
| "eval_runtime": 5.3199, | |
| "eval_samples_per_second": 13.158, | |
| "eval_steps_per_second": 0.94, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9571428571428572, | |
| "eval_loss": 0.21541821956634521, | |
| "eval_runtime": 5.3461, | |
| "eval_samples_per_second": 13.094, | |
| "eval_steps_per_second": 0.935, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 16.842105263157894, | |
| "grad_norm": 17.900920867919922, | |
| "learning_rate": 0.0, | |
| "loss": 0.0455, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 16.842105263157894, | |
| "eval_accuracy": 0.9428571428571428, | |
| "eval_loss": 0.2586957514286041, | |
| "eval_runtime": 5.2529, | |
| "eval_samples_per_second": 13.326, | |
| "eval_steps_per_second": 0.952, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 16.842105263157894, | |
| "step": 80, | |
| "total_flos": 3.991922735555543e+17, | |
| "train_loss": 0.10449823420494794, | |
| "train_runtime": 538.8903, | |
| "train_samples_per_second": 11.134, | |
| "train_steps_per_second": 0.148 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "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": 3.991922735555543e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |