last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00282585658777817,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00028258565877781704,
      "grad_norm": 7.329011917114258,
      "learning_rate": 2e-05,
      "loss": 4.3627,
      "step": 1
    },
    {
      "epoch": 0.00028258565877781704,
      "eval_loss": 1.151261806488037,
      "eval_runtime": 90.7506,
      "eval_samples_per_second": 16.419,
      "eval_steps_per_second": 8.209,
      "step": 1
    },
    {
      "epoch": 0.0005651713175556341,
      "grad_norm": 8.109373092651367,
      "learning_rate": 4e-05,
      "loss": 5.0664,
      "step": 2
    },
    {
      "epoch": 0.0008477569763334511,
      "grad_norm": 7.209464073181152,
      "learning_rate": 6e-05,
      "loss": 4.6849,
      "step": 3
    },
    {
      "epoch": 0.0008477569763334511,
      "eval_loss": 1.092017412185669,
      "eval_runtime": 84.8137,
      "eval_samples_per_second": 17.568,
      "eval_steps_per_second": 8.784,
      "step": 3
    },
    {
      "epoch": 0.0011303426351112682,
      "grad_norm": 6.724841594696045,
      "learning_rate": 8e-05,
      "loss": 4.5783,
      "step": 4
    },
    {
      "epoch": 0.001412928293889085,
      "grad_norm": 5.912840843200684,
      "learning_rate": 0.0001,
      "loss": 4.2418,
      "step": 5
    },
    {
      "epoch": 0.0016955139526669022,
      "grad_norm": 5.554118633270264,
      "learning_rate": 0.00012,
      "loss": 3.4454,
      "step": 6
    },
    {
      "epoch": 0.0016955139526669022,
      "eval_loss": 0.8012614846229553,
      "eval_runtime": 84.8502,
      "eval_samples_per_second": 17.56,
      "eval_steps_per_second": 8.78,
      "step": 6
    },
    {
      "epoch": 0.001978099611444719,
      "grad_norm": 6.197779655456543,
      "learning_rate": 0.00014,
      "loss": 3.307,
      "step": 7
    },
    {
      "epoch": 0.0022606852702225363,
      "grad_norm": 6.19174861907959,
      "learning_rate": 0.00016,
      "loss": 2.7262,
      "step": 8
    },
    {
      "epoch": 0.002543270929000353,
      "grad_norm": 5.924742698669434,
      "learning_rate": 0.00018,
      "loss": 2.5307,
      "step": 9
    },
    {
      "epoch": 0.002543270929000353,
      "eval_loss": 0.4889945685863495,
      "eval_runtime": 84.8789,
      "eval_samples_per_second": 17.554,
      "eval_steps_per_second": 8.777,
      "step": 9
    },
    {
      "epoch": 0.00282585658777817,
      "grad_norm": 8.411174774169922,
      "learning_rate": 0.0002,
      "loss": 1.7706,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3505343598428160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}