| { | |
| "best_metric": 0.9043070673942566, | |
| "best_model_checkpoint": "vit-base-cat-emotions\\checkpoint-200", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 1.1178287267684937, | |
| "learning_rate": 0.00019375000000000002, | |
| "loss": 1.9108, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 2.2460951805114746, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.815, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 1.9426186084747314, | |
| "learning_rate": 0.00018125000000000001, | |
| "loss": 1.5822, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.800794839859009, | |
| "learning_rate": 0.000175, | |
| "loss": 1.3129, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 3.274559497833252, | |
| "learning_rate": 0.00016875, | |
| "loss": 1.0768, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 2.4314801692962646, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 0.9329, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 1.755449891090393, | |
| "learning_rate": 0.00015625, | |
| "loss": 0.7257, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.522848129272461, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.519, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 1.7849164009094238, | |
| "learning_rate": 0.00014375, | |
| "loss": 0.4472, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 1.2824078798294067, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.3361, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "eval_accuracy": 0.6547619047619048, | |
| "eval_loss": 1.0124708414077759, | |
| "eval_runtime": 0.7491, | |
| "eval_samples_per_second": 112.131, | |
| "eval_steps_per_second": 14.684, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "grad_norm": 1.1555737257003784, | |
| "learning_rate": 0.00013125000000000002, | |
| "loss": 0.2081, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.7941139936447144, | |
| "learning_rate": 0.000125, | |
| "loss": 0.1453, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.39761993288993835, | |
| "learning_rate": 0.00011875, | |
| "loss": 0.1842, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.4260774254798889, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 0.0941, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 0.2251412719488144, | |
| "learning_rate": 0.00010625000000000001, | |
| "loss": 0.0809, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.22556072473526, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0827, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.175390362739563, | |
| "learning_rate": 9.375e-05, | |
| "loss": 0.0643, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 0.18123513460159302, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.0652, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.9375, | |
| "grad_norm": 0.15588343143463135, | |
| "learning_rate": 8.125000000000001e-05, | |
| "loss": 0.0498, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.2110791653394699, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0723, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_accuracy": 0.7380952380952381, | |
| "eval_loss": 0.9043070673942566, | |
| "eval_runtime": 0.7518, | |
| "eval_samples_per_second": 111.725, | |
| "eval_steps_per_second": 14.631, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.1404016613960266, | |
| "learning_rate": 6.875e-05, | |
| "loss": 0.0442, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.13061176240444183, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0529, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.1875, | |
| "grad_norm": 0.11943121254444122, | |
| "learning_rate": 5.6250000000000005e-05, | |
| "loss": 0.0381, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.11987230181694031, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0422, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.11512718349695206, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.0429, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 0.11094118654727936, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.034, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.4375, | |
| "grad_norm": 0.10723838955163956, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.0339, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.11790188401937485, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0425, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.0625, | |
| "grad_norm": 0.09828304499387741, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.0331, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.09926401823759079, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0321, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.926792323589325, | |
| "eval_runtime": 0.7517, | |
| "eval_samples_per_second": 111.752, | |
| "eval_steps_per_second": 14.634, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.6875, | |
| "grad_norm": 0.10707634687423706, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.0404, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.1315452754497528, | |
| "learning_rate": 0.0, | |
| "loss": 0.0319, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 320, | |
| "total_flos": 3.8902722072367104e+17, | |
| "train_loss": 0.3804276719689369, | |
| "train_runtime": 80.487, | |
| "train_samples_per_second": 62.37, | |
| "train_steps_per_second": 3.976 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 3.8902722072367104e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |