trainer_state.json

{
  "best_global_step": 200,
  "best_metric": 0.9849624060150376,
  "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": 1.961495280265808,
      "learning_rate": 0.0001930769230769231,
      "loss": 0.8212,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.9139995574951172,
      "learning_rate": 0.0001853846153846154,
      "loss": 0.2999,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.0324450731277466,
      "learning_rate": 0.0001776923076923077,
      "loss": 0.2349,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.7793644666671753,
      "learning_rate": 0.00017,
      "loss": 0.2267,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.23891262710094452,
      "learning_rate": 0.0001623076923076923,
      "loss": 0.1409,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 1.8466084003448486,
      "learning_rate": 0.00015461538461538464,
      "loss": 0.2156,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 6.694087028503418,
      "learning_rate": 0.00014692307692307693,
      "loss": 0.4367,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.2549947500228882,
      "learning_rate": 0.00013923076923076923,
      "loss": 0.1395,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.16331493854522705,
      "learning_rate": 0.00013153846153846156,
      "loss": 0.0647,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 6.384829998016357,
      "learning_rate": 0.00012384615384615385,
      "loss": 0.1103,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.09903952479362488,
      "eval_runtime": 2.1898,
      "eval_samples_per_second": 60.737,
      "eval_steps_per_second": 4.11,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 3.5876402854919434,
      "learning_rate": 0.00011615384615384617,
      "loss": 0.139,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.17117121815681458,
      "learning_rate": 0.00010846153846153846,
      "loss": 0.0855,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8414745330810547,
      "learning_rate": 0.00010076923076923077,
      "loss": 0.0644,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.25121811032295227,
      "learning_rate": 9.307692307692309e-05,
      "loss": 0.0337,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.15971487760543823,
      "learning_rate": 8.538461538461538e-05,
      "loss": 0.02,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.14554241299629211,
      "learning_rate": 7.76923076923077e-05,
      "loss": 0.0196,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.8151974081993103,
      "learning_rate": 7e-05,
      "loss": 0.0622,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.8853701949119568,
      "learning_rate": 6.23076923076923e-05,
      "loss": 0.0457,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.06993617862462997,
      "learning_rate": 5.461538461538461e-05,
      "loss": 0.0141,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.07229465246200562,
      "learning_rate": 4.692307692307693e-05,
      "loss": 0.0177,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.04772329330444336,
      "eval_runtime": 1.2784,
      "eval_samples_per_second": 104.039,
      "eval_steps_per_second": 7.04,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.06657698005437851,
      "learning_rate": 3.923076923076923e-05,
      "loss": 0.0127,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.06428707391023636,
      "learning_rate": 3.153846153846154e-05,
      "loss": 0.0124,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.06366737931966782,
      "learning_rate": 2.384615384615385e-05,
      "loss": 0.0118,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.05925724282860756,
      "learning_rate": 1.6153846153846154e-05,
      "loss": 0.0114,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.05729958042502403,
      "learning_rate": 8.461538461538462e-06,
      "loss": 0.0112,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05867352336645126,
      "learning_rate": 7.692307692307694e-07,
      "loss": 0.0112,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.12549979543456666,
      "train_runtime": 117.8808,
      "train_samples_per_second": 35.086,
      "train_steps_per_second": 2.206
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.205097416476426e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}