trainer_state.json

{
  "best_metric": 0.01428612694144249,
  "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": 1.690329670906067,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.86,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 3.2744600772857666,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3321,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.1044212579727173,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.1878,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.6217836141586304,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.1771,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.3753783702850342,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1713,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 6.028824329376221,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.3572,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.7344399690628052,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1494,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.15125687420368195,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.103,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.18919619917869568,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0585,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.7425695061683655,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0965,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.08969047665596008,
      "eval_runtime": 2.6147,
      "eval_samples_per_second": 50.867,
      "eval_steps_per_second": 6.502,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.3135984539985657,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1042,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.09811767935752869,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0379,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3020174503326416,
      "learning_rate": 0.0001,
      "loss": 0.0711,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.0861172154545784,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0203,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 2.5387868881225586,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0436,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 2.0208182334899902,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0755,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.07129581272602081,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0179,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.2694968283176422,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0197,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.06667916476726532,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0471,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06263082474470139,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0159,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01428612694144249,
      "eval_runtime": 2.3048,
      "eval_samples_per_second": 57.705,
      "eval_steps_per_second": 7.376,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.062446195632219315,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0124,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.05707734078168869,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0113,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.06064826622605324,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0338,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.05545338988304138,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0106,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.06174391508102417,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0241,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3382802903652191,
      "learning_rate": 0.0,
      "loss": 0.011,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11728346955317717,
      "train_runtime": 109.5895,
      "train_samples_per_second": 37.741,
      "train_steps_per_second": 2.372
    }
  ],
  "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
}