last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.012759170653907496,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004253056884635832,
      "eval_loss": NaN,
      "eval_runtime": 96.8528,
      "eval_samples_per_second": 10.222,
      "eval_steps_per_second": 5.111,
      "step": 1
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.002126528442317916,
      "eval_loss": NaN,
      "eval_runtime": 96.7385,
      "eval_samples_per_second": 10.234,
      "eval_steps_per_second": 5.117,
      "step": 5
    },
    {
      "epoch": 0.002551834130781499,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.003827751196172249,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.004253056884635832,
      "eval_loss": NaN,
      "eval_runtime": 96.7407,
      "eval_samples_per_second": 10.234,
      "eval_steps_per_second": 5.117,
      "step": 10
    },
    {
      "epoch": 0.005103668261562998,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.006379585326953748,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.006379585326953748,
      "eval_loss": NaN,
      "eval_runtime": 96.8062,
      "eval_samples_per_second": 10.227,
      "eval_steps_per_second": 5.113,
      "step": 15
    },
    {
      "epoch": 0.007655502392344498,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.008506113769271665,
      "eval_loss": NaN,
      "eval_runtime": 96.7895,
      "eval_samples_per_second": 10.228,
      "eval_steps_per_second": 5.114,
      "step": 20
    },
    {
      "epoch": 0.008931419457735247,
      "grad_norm": NaN,
      "learning_rate": 4.2178276747988446e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.010207336523125997,
      "grad_norm": NaN,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.01063264221158958,
      "eval_loss": NaN,
      "eval_runtime": 96.8128,
      "eval_samples_per_second": 10.226,
      "eval_steps_per_second": 5.113,
      "step": 25
    },
    {
      "epoch": 0.011483253588516746,
      "grad_norm": NaN,
      "learning_rate": 5.449673790581611e-06,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.012759170653907496,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.012759170653907496,
      "eval_loss": NaN,
      "eval_runtime": 96.7972,
      "eval_samples_per_second": 10.228,
      "eval_steps_per_second": 5.114,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.907240949566669e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}