last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003727831054696601,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.727831054696601e-05,
      "grad_norm": 10.861333847045898,
      "learning_rate": 2e-05,
      "loss": 13.1054,
      "step": 1
    },
    {
      "epoch": 3.727831054696601e-05,
      "eval_loss": 9.946808815002441,
      "eval_runtime": 508.0043,
      "eval_samples_per_second": 22.234,
      "eval_steps_per_second": 11.118,
      "step": 1
    },
    {
      "epoch": 7.455662109393202e-05,
      "grad_norm": 6.481636047363281,
      "learning_rate": 4e-05,
      "loss": 9.4534,
      "step": 2
    },
    {
      "epoch": 0.00011183493164089804,
      "grad_norm": 17.837316513061523,
      "learning_rate": 6e-05,
      "loss": 11.985,
      "step": 3
    },
    {
      "epoch": 0.00011183493164089804,
      "eval_loss": 9.895621299743652,
      "eval_runtime": 506.0864,
      "eval_samples_per_second": 22.318,
      "eval_steps_per_second": 11.16,
      "step": 3
    },
    {
      "epoch": 0.00014911324218786404,
      "grad_norm": 4.463501930236816,
      "learning_rate": 8e-05,
      "loss": 5.4877,
      "step": 4
    },
    {
      "epoch": 0.00018639155273483004,
      "grad_norm": 7.662845611572266,
      "learning_rate": 0.0001,
      "loss": 7.9353,
      "step": 5
    },
    {
      "epoch": 0.00022366986328179608,
      "grad_norm": 5.951407432556152,
      "learning_rate": 0.00012,
      "loss": 5.9694,
      "step": 6
    },
    {
      "epoch": 0.00022366986328179608,
      "eval_loss": 8.601508140563965,
      "eval_runtime": 505.7711,
      "eval_samples_per_second": 22.332,
      "eval_steps_per_second": 11.167,
      "step": 6
    },
    {
      "epoch": 0.0002609481738287621,
      "grad_norm": 14.010904312133789,
      "learning_rate": 0.00014,
      "loss": 9.0872,
      "step": 7
    },
    {
      "epoch": 0.00029822648437572807,
      "grad_norm": 7.74440336227417,
      "learning_rate": 0.00016,
      "loss": 4.2274,
      "step": 8
    },
    {
      "epoch": 0.0003355047949226941,
      "grad_norm": 15.858744621276855,
      "learning_rate": 0.00018,
      "loss": 7.5469,
      "step": 9
    },
    {
      "epoch": 0.0003355047949226941,
      "eval_loss": 4.148334503173828,
      "eval_runtime": 506.2852,
      "eval_samples_per_second": 22.31,
      "eval_steps_per_second": 11.156,
      "step": 9
    },
    {
      "epoch": 0.0003727831054696601,
      "grad_norm": 10.15648078918457,
      "learning_rate": 0.0002,
      "loss": 4.5332,
      "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": 2820585478422528.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}