last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0002702580964821404,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.0028677386904494e-05,
      "grad_norm": 0.2012191116809845,
      "learning_rate": 2e-05,
      "loss": 0.3476,
      "step": 1
    },
    {
      "epoch": 3.0028677386904494e-05,
      "eval_loss": 0.3906751573085785,
      "eval_runtime": 1045.4932,
      "eval_samples_per_second": 13.412,
      "eval_steps_per_second": 6.706,
      "step": 1
    },
    {
      "epoch": 6.005735477380899e-05,
      "grad_norm": 0.3361751139163971,
      "learning_rate": 4e-05,
      "loss": 0.5332,
      "step": 2
    },
    {
      "epoch": 9.008603216071348e-05,
      "grad_norm": 0.16664037108421326,
      "learning_rate": 6e-05,
      "loss": 0.282,
      "step": 3
    },
    {
      "epoch": 9.008603216071348e-05,
      "eval_loss": 0.38874995708465576,
      "eval_runtime": 1044.056,
      "eval_samples_per_second": 13.43,
      "eval_steps_per_second": 6.715,
      "step": 3
    },
    {
      "epoch": 0.00012011470954761798,
      "grad_norm": 0.283048540353775,
      "learning_rate": 8e-05,
      "loss": 0.3772,
      "step": 4
    },
    {
      "epoch": 0.00015014338693452247,
      "grad_norm": 0.6533939838409424,
      "learning_rate": 0.0001,
      "loss": 0.3009,
      "step": 5
    },
    {
      "epoch": 0.00018017206432142696,
      "grad_norm": 0.32610440254211426,
      "learning_rate": 0.00012,
      "loss": 0.4086,
      "step": 6
    },
    {
      "epoch": 0.00018017206432142696,
      "eval_loss": 0.3601542115211487,
      "eval_runtime": 1044.8592,
      "eval_samples_per_second": 13.42,
      "eval_steps_per_second": 6.71,
      "step": 6
    },
    {
      "epoch": 0.00021020074170833146,
      "grad_norm": 0.5947815179824829,
      "learning_rate": 0.00014,
      "loss": 0.3591,
      "step": 7
    },
    {
      "epoch": 0.00024022941909523595,
      "grad_norm": 0.3068762719631195,
      "learning_rate": 0.00016,
      "loss": 0.2623,
      "step": 8
    },
    {
      "epoch": 0.0002702580964821404,
      "grad_norm": 0.5450010895729065,
      "learning_rate": 0.00018,
      "loss": 0.2086,
      "step": 9
    },
    {
      "epoch": 0.0002702580964821404,
      "eval_loss": 0.23924924433231354,
      "eval_runtime": 1044.6169,
      "eval_samples_per_second": 13.423,
      "eval_steps_per_second": 6.712,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4986771571998720.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}