checkpoint-300/trainer_state.json

{
  "best_metric": 0.5696164965629578,
  "best_model_checkpoint": "/kaggle/output/checkpoint-100",
  "epoch": 0.6198347107438017,
  "eval_steps": 50,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0,
      "loss": 0.7246,
      "step": 1
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.333333333333332e-05,
      "loss": 0.5982,
      "step": 50
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6296697854995728,
      "eval_runtime": 30.5612,
      "eval_samples_per_second": 38.742,
      "eval_steps_per_second": 19.371,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00015,
      "loss": 0.5505,
      "step": 100
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5696164965629578,
      "eval_runtime": 30.5816,
      "eval_samples_per_second": 38.716,
      "eval_steps_per_second": 19.358,
      "step": 100
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002333333333333333,
      "loss": 0.5838,
      "step": 150
    },
    {
      "epoch": 0.31,
      "eval_accuracy": 0.75,
      "eval_loss": 0.562861442565918,
      "eval_runtime": 31.0771,
      "eval_samples_per_second": 38.099,
      "eval_steps_per_second": 19.049,
      "step": 150
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00025333333333333333,
      "loss": 0.5925,
      "step": 200
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5931239724159241,
      "eval_runtime": 30.6023,
      "eval_samples_per_second": 38.69,
      "eval_steps_per_second": 19.345,
      "step": 200
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00025333333333333333,
      "loss": 0.7003,
      "step": 250
    },
    {
      "epoch": 0.52,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5931239724159241,
      "eval_runtime": 30.7477,
      "eval_samples_per_second": 38.507,
      "eval_steps_per_second": 19.253,
      "step": 250
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00025333333333333333,
      "loss": 0.606,
      "step": 300
    },
    {
      "epoch": 0.62,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5931239724159241,
      "eval_runtime": 30.6306,
      "eval_samples_per_second": 38.654,
      "eval_steps_per_second": 19.327,
      "step": 300
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1043244793555752.0,
  "trial_name": null,
  "trial_params": null
}