checkpoint-150/trainer_state.json

{
 "best_metric": 1.0,
 "best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-50",
 "epoch": 2.25,
 "eval_steps": 500,
 "global_step": 150,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.05,
 "grad_norm": 14.361767768859863,
 "learning_rate": 2.5e-05,
 "loss": 2.3816,
 "step": 10
 },
 {
 "epoch": 0.1,
 "grad_norm": 8.448980331420898,
 "learning_rate": 5e-05,
 "loss": 1.5686,
 "step": 20
 },
 {
 "epoch": 0.15,
 "grad_norm": 7.0012526512146,
 "learning_rate": 4.722222222222222e-05,
 "loss": 0.6191,
 "step": 30
 },
 {
 "epoch": 0.2,
 "grad_norm": 4.961479187011719,
 "learning_rate": 4.4444444444444447e-05,
 "loss": 0.3468,
 "step": 40
 },
 {
 "epoch": 0.25,
 "grad_norm": 2.13403058052063,
 "learning_rate": 4.166666666666667e-05,
 "loss": 0.1523,
 "step": 50
 },
 {
 "epoch": 0.25,
 "eval_accuracy": 1.0,
 "eval_loss": 0.07354684174060822,
 "eval_runtime": 34.9344,
 "eval_samples_per_second": 4.265,
 "eval_steps_per_second": 0.716,
 "step": 50
 },
 {
 "epoch": 1.05,
 "grad_norm": 0.933645486831665,
 "learning_rate": 3.888888888888889e-05,
 "loss": 0.0521,
 "step": 60
 },
 {
 "epoch": 1.1,
 "grad_norm": 0.865109920501709,
 "learning_rate": 3.611111111111111e-05,
 "loss": 0.0226,
 "step": 70
 },
 {
 "epoch": 1.15,
 "grad_norm": 0.17869924008846283,
 "learning_rate": 3.3333333333333335e-05,
 "loss": 0.108,
 "step": 80
 },
 {
 "epoch": 1.2,
 "grad_norm": 0.23531177639961243,
 "learning_rate": 3.055555555555556e-05,
 "loss": 0.0107,
 "step": 90
 },
 {
 "epoch": 1.25,
 "grad_norm": 0.1640271246433258,
 "learning_rate": 2.777777777777778e-05,
 "loss": 0.024,
 "step": 100
 },
 {
 "epoch": 1.25,
 "eval_accuracy": 0.9865771812080537,
 "eval_loss": 0.046999599784612656,
 "eval_runtime": 32.3832,
 "eval_samples_per_second": 4.601,
 "eval_steps_per_second": 0.772,
 "step": 100
 },
 {
 "epoch": 2.05,
 "grad_norm": 0.47139155864715576,
 "learning_rate": 2.5e-05,
 "loss": 0.0056,
 "step": 110
 },
 {
 "epoch": 2.1,
 "grad_norm": 0.11248086392879486,
 "learning_rate": 2.2222222222222223e-05,
 "loss": 0.0045,
 "step": 120
 },
 {
 "epoch": 2.15,
 "grad_norm": 0.07324650138616562,
 "learning_rate": 1.9444444444444445e-05,
 "loss": 0.0042,
 "step": 130
 },
 {
 "epoch": 2.2,
 "grad_norm": 0.46895766258239746,
 "learning_rate": 1.6666666666666667e-05,
 "loss": 0.0202,
 "step": 140
 },
 {
 "epoch": 2.25,
 "grad_norm": 0.048765964806079865,
 "learning_rate": 1.388888888888889e-05,
 "loss": 0.0583,
 "step": 150
 },
 {
 "epoch": 2.25,
 "eval_accuracy": 0.9865771812080537,
 "eval_loss": 0.030154313892126083,
 "eval_runtime": 32.7398,
 "eval_samples_per_second": 4.551,
 "eval_steps_per_second": 0.764,
 "step": 150
 }
 ],
 "logging_steps": 10,
 "max_steps": 200,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 9223372036854775807,
 "save_steps": 500,
 "total_flos": 7.885725831954432e+17,
 "train_batch_size": 6,
 "trial_name": null,
 "trial_params": null
}