trainer_state.json

{
  "best_metric": 0.5925925925925926,
  "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-conspiracy_imagery_2/checkpoint-76",
  "epoch": 5.901639344262295,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 6.740944862365723,
      "learning_rate": 4.938271604938271e-05,
      "loss": 1.9515,
      "step": 10
    },
    {
      "epoch": 0.9836065573770492,
      "eval_accuracy": 0.42592592592592593,
      "eval_loss": 1.3559091091156006,
      "eval_runtime": 3.4851,
      "eval_samples_per_second": 30.989,
      "eval_steps_per_second": 2.009,
      "step": 15
    },
    {
      "epoch": 1.3114754098360657,
      "grad_norm": 6.367587566375732,
      "learning_rate": 4.3209876543209875e-05,
      "loss": 1.3689,
      "step": 20
    },
    {
      "epoch": 1.9672131147540983,
      "grad_norm": 6.214643478393555,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 1.1593,
      "step": 30
    },
    {
      "epoch": 1.9672131147540983,
      "eval_accuracy": 0.5648148148148148,
      "eval_loss": 1.1501442193984985,
      "eval_runtime": 3.5137,
      "eval_samples_per_second": 30.737,
      "eval_steps_per_second": 1.992,
      "step": 30
    },
    {
      "epoch": 2.6229508196721314,
      "grad_norm": 5.606908798217773,
      "learning_rate": 3.08641975308642e-05,
      "loss": 1.0288,
      "step": 40
    },
    {
      "epoch": 2.9508196721311473,
      "eval_accuracy": 0.5462962962962963,
      "eval_loss": 1.1120755672454834,
      "eval_runtime": 3.3763,
      "eval_samples_per_second": 31.987,
      "eval_steps_per_second": 2.073,
      "step": 45
    },
    {
      "epoch": 3.278688524590164,
      "grad_norm": 5.3310627937316895,
      "learning_rate": 2.4691358024691357e-05,
      "loss": 0.9132,
      "step": 50
    },
    {
      "epoch": 3.9344262295081966,
      "grad_norm": 5.54246187210083,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.8302,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5740740740740741,
      "eval_loss": 1.1271188259124756,
      "eval_runtime": 3.3391,
      "eval_samples_per_second": 32.344,
      "eval_steps_per_second": 2.096,
      "step": 61
    },
    {
      "epoch": 4.590163934426229,
      "grad_norm": 5.834253787994385,
      "learning_rate": 1.2345679012345678e-05,
      "loss": 0.7957,
      "step": 70
    },
    {
      "epoch": 4.983606557377049,
      "eval_accuracy": 0.5925925925925926,
      "eval_loss": 1.0559544563293457,
      "eval_runtime": 4.0977,
      "eval_samples_per_second": 26.356,
      "eval_steps_per_second": 1.708,
      "step": 76
    },
    {
      "epoch": 5.245901639344262,
      "grad_norm": 6.654218673706055,
      "learning_rate": 6.172839506172839e-06,
      "loss": 0.8094,
      "step": 80
    },
    {
      "epoch": 5.901639344262295,
      "grad_norm": 7.636894702911377,
      "learning_rate": 0.0,
      "loss": 0.7633,
      "step": 90
    },
    {
      "epoch": 5.901639344262295,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.08096182346344,
      "eval_runtime": 3.983,
      "eval_samples_per_second": 27.115,
      "eval_steps_per_second": 1.757,
      "step": 90
    },
    {
      "epoch": 5.901639344262295,
      "step": 90,
      "total_flos": 4.4386319544827904e+17,
      "train_loss": 1.0689217143588596,
      "train_runtime": 507.807,
      "train_samples_per_second": 11.461,
      "train_steps_per_second": 0.177
    }
  ],
  "logging_steps": 10,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.4386319544827904e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}