trainer_state.json

{
  "best_metric": 0.015216110274195671,
  "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": 1.5747699737548828,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8196,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.289270043373108,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3579,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.0927575826644897,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2194,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.846202552318573,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2587,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.3544741868972778,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1994,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 3.535122871398926,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2447,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 3.648043394088745,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.0984,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.1548488438129425,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0793,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 3.230766534805298,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0905,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.13816161453723907,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0809,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.056199491024017334,
      "eval_runtime": 1.2935,
      "eval_samples_per_second": 102.825,
      "eval_steps_per_second": 13.143,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 1.687045931816101,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0756,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 4.99824857711792,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0797,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.942594051361084,
      "learning_rate": 0.0001,
      "loss": 0.0826,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.6112641096115112,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0688,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.09129239618778229,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0425,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.14322307705879211,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0893,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.0938325896859169,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0217,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 3.1701836585998535,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0348,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 10.15375804901123,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0622,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.0709332823753357,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0296,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.015216110274195671,
      "eval_runtime": 1.2552,
      "eval_samples_per_second": 105.955,
      "eval_steps_per_second": 13.543,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.6941270232200623,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0151,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.21077759563922882,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0132,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.2554996907711029,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0126,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.06735586374998093,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0121,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.06165779009461403,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.012,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05904724821448326,
      "learning_rate": 0.0,
      "loss": 0.0116,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11970446820442493,
      "train_runtime": 92.2528,
      "train_samples_per_second": 44.833,
      "train_steps_per_second": 2.818
    }
  ],
  "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
}