trainer_state.json

{
  "best_metric": 0.9361702127659575,
  "best_model_checkpoint": "portrait_cosu_exp3/checkpoint-13",
  "epoch": 3.6923076923076925,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9230769230769231,
      "eval_accuracy": 0.851063829787234,
      "eval_confusion_matrix": [
        [
          17,
          3
        ],
        [
          4,
          23
        ]
      ],
      "eval_f1": 0.8514750663363719,
      "eval_loss": 0.2920527458190918,
      "eval_precision": 0.8526615228742889,
      "eval_recall": 0.851063829787234,
      "eval_runtime": 3.0689,
      "eval_samples_per_second": 15.315,
      "eval_steps_per_second": 0.978,
      "step": 6
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 7.153378486633301,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.5415,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9361702127659575,
      "eval_confusion_matrix": [
        [
          17,
          3
        ],
        [
          0,
          27
        ]
      ],
      "eval_f1": 0.9352622499319029,
      "eval_loss": 0.25635045766830444,
      "eval_precision": 0.9425531914893617,
      "eval_recall": 0.9361702127659575,
      "eval_runtime": 3.2937,
      "eval_samples_per_second": 14.27,
      "eval_steps_per_second": 0.911,
      "step": 13
    },
    {
      "epoch": 2.9230769230769234,
      "eval_accuracy": 0.8723404255319149,
      "eval_confusion_matrix": [
        [
          19,
          1
        ],
        [
          5,
          22
        ]
      ],
      "eval_f1": 0.873036750483559,
      "eval_loss": 0.3604692220687866,
      "eval_precision": 0.8863706444650015,
      "eval_recall": 0.8723404255319149,
      "eval_runtime": 3.1924,
      "eval_samples_per_second": 14.722,
      "eval_steps_per_second": 0.94,
      "step": 19
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 9.893902778625488,
      "learning_rate": 9.523809523809523e-06,
      "loss": 0.378,
      "step": 20
    },
    {
      "epoch": 3.6923076923076925,
      "eval_accuracy": 0.9148936170212766,
      "eval_confusion_matrix": [
        [
          19,
          1
        ],
        [
          3,
          24
        ]
      ],
      "eval_f1": 0.9152832982620216,
      "eval_loss": 0.25239235162734985,
      "eval_precision": 0.9189941972920695,
      "eval_recall": 0.9148936170212766,
      "eval_runtime": 2.9295,
      "eval_samples_per_second": 16.044,
      "eval_steps_per_second": 1.024,
      "step": 24
    },
    {
      "epoch": 3.6923076923076925,
      "step": 24,
      "total_flos": 3.505150325906473e+17,
      "train_loss": 0.428266316652298,
      "train_runtime": 301.1045,
      "train_samples_per_second": 5.526,
      "train_steps_per_second": 0.08
    }
  ],
  "logging_steps": 10,
  "max_steps": 24,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 3.505150325906473e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}