last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.097560975609756,
  "eval_steps": 200,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.1872,
      "step": 200
    },
    {
      "epoch": 0.61,
      "eval_runtime": 75.0556,
      "eval_samples_per_second": 3.677,
      "eval_steps_per_second": 0.24,
      "step": 200
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.0954,
      "step": 400
    },
    {
      "epoch": 1.22,
      "eval_runtime": 74.9815,
      "eval_samples_per_second": 3.681,
      "eval_steps_per_second": 0.24,
      "step": 400
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.0797,
      "step": 600
    },
    {
      "epoch": 1.83,
      "eval_runtime": 75.1061,
      "eval_samples_per_second": 3.675,
      "eval_steps_per_second": 0.24,
      "step": 600
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0002,
      "loss": 0.06,
      "step": 800
    },
    {
      "epoch": 2.44,
      "eval_runtime": 74.9943,
      "eval_samples_per_second": 3.68,
      "eval_steps_per_second": 0.24,
      "step": 800
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0002,
      "loss": 0.0551,
      "step": 1000
    },
    {
      "epoch": 3.05,
      "eval_runtime": 74.9839,
      "eval_samples_per_second": 3.681,
      "eval_steps_per_second": 0.24,
      "step": 1000
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.0002,
      "loss": 0.0422,
      "step": 1200
    },
    {
      "epoch": 3.66,
      "eval_runtime": 75.0437,
      "eval_samples_per_second": 3.678,
      "eval_steps_per_second": 0.24,
      "step": 1200
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0002,
      "loss": 0.041,
      "step": 1400
    },
    {
      "epoch": 4.27,
      "eval_runtime": 74.9611,
      "eval_samples_per_second": 3.682,
      "eval_steps_per_second": 0.24,
      "step": 1400
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.0002,
      "loss": 0.038,
      "step": 1600
    },
    {
      "epoch": 4.88,
      "eval_runtime": 75.1592,
      "eval_samples_per_second": 3.672,
      "eval_steps_per_second": 0.239,
      "step": 1600
    },
    {
      "epoch": 5.49,
      "learning_rate": 0.0002,
      "loss": 0.0348,
      "step": 1800
    },
    {
      "epoch": 5.49,
      "eval_runtime": 75.1286,
      "eval_samples_per_second": 3.674,
      "eval_steps_per_second": 0.24,
      "step": 1800
    },
    {
      "epoch": 6.1,
      "learning_rate": 0.0002,
      "loss": 0.0366,
      "step": 2000
    },
    {
      "epoch": 6.1,
      "eval_runtime": 34.8655,
      "eval_samples_per_second": 7.916,
      "eval_steps_per_second": 0.516,
      "step": 2000
    }
  ],
  "logging_steps": 200,
  "max_steps": 3936,
  "num_train_epochs": 12,
  "save_steps": 200,
  "total_flos": 6.442331252195328e+17,
  "trial_name": null,
  "trial_params": null
}