trainer_state.json

{
  "best_metric": 0.02210470847785473,
  "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": 1.4500569105148315,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8762,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.141402006149292,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3944,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 4.175230979919434,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2569,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.4067094326019287,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.3333,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.4498758912086487,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2024,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 3.618130922317505,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2785,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.37593412399292,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1306,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 3.8326995372772217,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1567,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 4.000864505767822,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.1011,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.2737503051757812,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0644,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.11050426214933395,
      "eval_runtime": 3.1522,
      "eval_samples_per_second": 42.193,
      "eval_steps_per_second": 5.393,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 6.804135799407959,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.13,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 4.175411701202393,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1627,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.11167668551206589,
      "learning_rate": 0.0001,
      "loss": 0.0552,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.0995037779211998,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0299,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.12060802429914474,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0188,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.0828489288687706,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0319,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.15137769281864166,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0403,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.11208732426166534,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.04,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.06881953030824661,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0291,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06956042349338531,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0436,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.02210470847785473,
      "eval_runtime": 1.2431,
      "eval_samples_per_second": 106.989,
      "eval_steps_per_second": 13.675,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.065009705722332,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0135,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.06288763880729675,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.015,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.0701344683766365,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0123,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.062173016369342804,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0128,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.06328325718641281,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0121,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05732400342822075,
      "learning_rate": 0.0,
      "loss": 0.0116,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.13281054170085835,
      "train_runtime": 85.7697,
      "train_samples_per_second": 48.222,
      "train_steps_per_second": 3.031
    }
  ],
  "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
}