trainer_state.json

{
  "best_metric": 0.0528901182115078,
  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 260,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.5694985389709473,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.779,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 3.1738662719726562,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3121,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.6277250051498413,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2458,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.249167799949646,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2304,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.7504239082336426,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1661,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 1.7937676906585693,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2717,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 2.10583758354187,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1335,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.16493375599384308,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1085,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 3.5142745971679688,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0776,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 7.79476261138916,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0504,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.11275270581245422,
      "eval_runtime": 1.2651,
      "eval_samples_per_second": 105.131,
      "eval_steps_per_second": 13.438,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 1.4835689067840576,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0954,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.34010475873947144,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1372,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2076033502817154,
      "learning_rate": 0.0001,
      "loss": 0.0853,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.09932549297809601,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0269,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.1113138496875763,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0717,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.4321606755256653,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0429,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.18629103899002075,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0227,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.070466049015522,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0465,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.3304802179336548,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0408,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06198474392294884,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0308,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.0528901182115078,
      "eval_runtime": 1.8145,
      "eval_samples_per_second": 73.297,
      "eval_steps_per_second": 9.369,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.08140137046575546,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0121,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.05991177260875702,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0117,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.06039084494113922,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0114,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.05764195695519447,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0118,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.05695033073425293,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.011,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05573964864015579,
      "learning_rate": 0.0,
      "loss": 0.0138,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11719392985105515,
      "train_runtime": 105.1308,
      "train_samples_per_second": 39.341,
      "train_steps_per_second": 2.473
    }
  ],
  "logging_steps": 10,
  "max_steps": 260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.205097416476426e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}