trainer_state.json

{
  "best_metric": 0.359375,
  "best_model_checkpoint": "finetuned-Accident-MultipleLabels-Video-subset-v2/checkpoint-12",
  "epoch": 12.04,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "eval_accuracy": 0.015625,
      "eval_loss": 2.055873394012451,
      "eval_runtime": 168.1651,
      "eval_samples_per_second": 0.381,
      "eval_steps_per_second": 0.095,
      "step": 4
    },
    {
      "epoch": 1.08,
      "eval_accuracy": 0.171875,
      "eval_loss": 1.9961313009262085,
      "eval_runtime": 166.2447,
      "eval_samples_per_second": 0.385,
      "eval_steps_per_second": 0.096,
      "step": 8
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.7392,
      "step": 10
    },
    {
      "epoch": 2.08,
      "eval_accuracy": 0.359375,
      "eval_loss": 2.103482246398926,
      "eval_runtime": 165.284,
      "eval_samples_per_second": 0.387,
      "eval_steps_per_second": 0.097,
      "step": 12
    },
    {
      "epoch": 3.08,
      "eval_accuracy": 0.34375,
      "eval_loss": 2.089285373687744,
      "eval_runtime": 165.3216,
      "eval_samples_per_second": 0.387,
      "eval_steps_per_second": 0.097,
      "step": 16
    },
    {
      "epoch": 4.08,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.4679,
      "step": 20
    },
    {
      "epoch": 4.08,
      "eval_accuracy": 0.203125,
      "eval_loss": 2.0624852180480957,
      "eval_runtime": 166.3144,
      "eval_samples_per_second": 0.385,
      "eval_steps_per_second": 0.096,
      "step": 20
    },
    {
      "epoch": 5.08,
      "eval_accuracy": 0.3125,
      "eval_loss": 1.9844822883605957,
      "eval_runtime": 165.6444,
      "eval_samples_per_second": 0.386,
      "eval_steps_per_second": 0.097,
      "step": 24
    },
    {
      "epoch": 6.08,
      "eval_accuracy": 0.3125,
      "eval_loss": 1.982945203781128,
      "eval_runtime": 165.7415,
      "eval_samples_per_second": 0.386,
      "eval_steps_per_second": 0.097,
      "step": 28
    },
    {
      "epoch": 7.04,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 1.3757,
      "step": 30
    },
    {
      "epoch": 7.08,
      "eval_accuracy": 0.34375,
      "eval_loss": 1.9941210746765137,
      "eval_runtime": 165.4328,
      "eval_samples_per_second": 0.387,
      "eval_steps_per_second": 0.097,
      "step": 32
    },
    {
      "epoch": 8.08,
      "eval_accuracy": 0.3125,
      "eval_loss": 2.0448787212371826,
      "eval_runtime": 165.4589,
      "eval_samples_per_second": 0.387,
      "eval_steps_per_second": 0.097,
      "step": 36
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 1.2956,
      "step": 40
    },
    {
      "epoch": 9.08,
      "eval_accuracy": 0.328125,
      "eval_loss": 2.087737560272217,
      "eval_runtime": 164.0412,
      "eval_samples_per_second": 0.39,
      "eval_steps_per_second": 0.098,
      "step": 40
    },
    {
      "epoch": 10.08,
      "eval_accuracy": 0.328125,
      "eval_loss": 2.122446060180664,
      "eval_runtime": 165.3005,
      "eval_samples_per_second": 0.387,
      "eval_steps_per_second": 0.097,
      "step": 44
    },
    {
      "epoch": 11.08,
      "eval_accuracy": 0.328125,
      "eval_loss": 2.141845703125,
      "eval_runtime": 164.5983,
      "eval_samples_per_second": 0.389,
      "eval_steps_per_second": 0.097,
      "step": 48
    },
    {
      "epoch": 12.04,
      "learning_rate": 0.0,
      "loss": 1.2352,
      "step": 50
    },
    {
      "epoch": 12.04,
      "eval_accuracy": 0.328125,
      "eval_loss": 2.147944211959839,
      "eval_runtime": 166.0407,
      "eval_samples_per_second": 0.385,
      "eval_steps_per_second": 0.096,
      "step": 50
    },
    {
      "epoch": 12.04,
      "step": 50,
      "total_flos": 9.968961218936832e+17,
      "train_loss": 1.422707748413086,
      "train_runtime": 4312.0314,
      "train_samples_per_second": 0.186,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 10,
  "max_steps": 50,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 9.968961218936832e+17,
  "trial_name": null,
  "trial_params": null
}