{
  "best_metric": 0.32018494606018066,
  "best_model_checkpoint": "facial_emotions_image_detection/checkpoint-8260",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 8260,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.42,
      "grad_norm": 13.829694747924805,
      "learning_rate": 2.856230031948882e-06,
      "loss": 0.399,
      "step": 500
    },
    {
      "epoch": 0.85,
      "grad_norm": 10.43336296081543,
      "learning_rate": 2.696485623003195e-06,
      "loss": 0.3907,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8980532379817243,
      "eval_loss": 0.331790030002594,
      "eval_runtime": 228.9339,
      "eval_samples_per_second": 109.944,
      "eval_steps_per_second": 13.746,
      "step": 1180
    },
    {
      "epoch": 1.27,
      "grad_norm": 13.529434204101562,
      "learning_rate": 2.536741214057508e-06,
      "loss": 0.3646,
      "step": 1500
    },
    {
      "epoch": 1.69,
      "grad_norm": 13.766672134399414,
      "learning_rate": 2.376996805111821e-06,
      "loss": 0.3674,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8970997218911403,
      "eval_loss": 0.33308491110801697,
      "eval_runtime": 226.1763,
      "eval_samples_per_second": 111.285,
      "eval_steps_per_second": 13.914,
      "step": 2360
    },
    {
      "epoch": 2.12,
      "grad_norm": 15.129395484924316,
      "learning_rate": 2.217252396166134e-06,
      "loss": 0.3738,
      "step": 2500
    },
    {
      "epoch": 2.54,
      "grad_norm": 15.538688659667969,
      "learning_rate": 2.057507987220447e-06,
      "loss": 0.3534,
      "step": 3000
    },
    {
      "epoch": 2.97,
      "grad_norm": 15.868049621582031,
      "learning_rate": 1.8977635782747604e-06,
      "loss": 0.3467,
      "step": 3500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8961064759634485,
      "eval_loss": 0.33278992772102356,
      "eval_runtime": 228.0713,
      "eval_samples_per_second": 110.36,
      "eval_steps_per_second": 13.798,
      "step": 3540
    },
    {
      "epoch": 3.39,
      "grad_norm": 12.208673477172852,
      "learning_rate": 1.7380191693290735e-06,
      "loss": 0.3321,
      "step": 4000
    },
    {
      "epoch": 3.81,
      "grad_norm": 12.523385047912598,
      "learning_rate": 1.5782747603833867e-06,
      "loss": 0.322,
      "step": 4500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8984505363528009,
      "eval_loss": 0.3272715210914612,
      "eval_runtime": 226.7446,
      "eval_samples_per_second": 111.006,
      "eval_steps_per_second": 13.879,
      "step": 4720
    },
    {
      "epoch": 4.24,
      "grad_norm": 15.311544418334961,
      "learning_rate": 1.4185303514376998e-06,
      "loss": 0.3288,
      "step": 5000
    },
    {
      "epoch": 4.66,
      "grad_norm": 13.10958194732666,
      "learning_rate": 1.2587859424920128e-06,
      "loss": 0.3182,
      "step": 5500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9007548669050457,
      "eval_loss": 0.3223397731781006,
      "eval_runtime": 225.1518,
      "eval_samples_per_second": 111.791,
      "eval_steps_per_second": 13.977,
      "step": 5900
    },
    {
      "epoch": 5.08,
      "grad_norm": 10.398481369018555,
      "learning_rate": 1.0990415335463259e-06,
      "loss": 0.2982,
      "step": 6000
    },
    {
      "epoch": 5.51,
      "grad_norm": 20.114734649658203,
      "learning_rate": 9.39297124600639e-07,
      "loss": 0.2946,
      "step": 6500
    },
    {
      "epoch": 5.93,
      "grad_norm": 20.724897384643555,
      "learning_rate": 7.795527156549522e-07,
      "loss": 0.3081,
      "step": 7000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9009137862534764,
      "eval_loss": 0.3215692341327667,
      "eval_runtime": 224.6793,
      "eval_samples_per_second": 112.026,
      "eval_steps_per_second": 14.007,
      "step": 7080
    },
    {
      "epoch": 6.36,
      "grad_norm": 13.87393856048584,
      "learning_rate": 6.198083067092652e-07,
      "loss": 0.296,
      "step": 7500
    },
    {
      "epoch": 6.78,
      "grad_norm": 8.323355674743652,
      "learning_rate": 4.600638977635783e-07,
      "loss": 0.2816,
      "step": 8000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9007945967421533,
      "eval_loss": 0.32018494606018066,
      "eval_runtime": 228.3984,
      "eval_samples_per_second": 110.202,
      "eval_steps_per_second": 13.779,
      "step": 8260
    }
  ],
  "logging_steps": 500,
  "max_steps": 9440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 2.047980331630782e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}