trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.625,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.025,
      "grad_norm": 135.99700927734375,
      "learning_rate": 0.0,
      "loss": 1.8567,
      "step": 1
    },
    {
      "epoch": 0.05,
      "grad_norm": 132.19590759277344,
      "learning_rate": 1e-06,
      "loss": 1.6451,
      "step": 2
    },
    {
      "epoch": 0.075,
      "grad_norm": 126.63935852050781,
      "learning_rate": 1e-06,
      "loss": 1.6308,
      "step": 3
    },
    {
      "epoch": 0.1,
      "grad_norm": 132.1169891357422,
      "learning_rate": 1e-06,
      "loss": 1.1376,
      "step": 4
    },
    {
      "epoch": 0.125,
      "grad_norm": 111.94847869873047,
      "learning_rate": 1e-06,
      "loss": 0.9303,
      "step": 5
    },
    {
      "epoch": 0.15,
      "grad_norm": 47.188209533691406,
      "learning_rate": 1e-06,
      "loss": 0.5403,
      "step": 6
    },
    {
      "epoch": 0.175,
      "grad_norm": 32.26974868774414,
      "learning_rate": 1e-06,
      "loss": 0.4366,
      "step": 7
    },
    {
      "epoch": 0.2,
      "grad_norm": 26.274761199951172,
      "learning_rate": 1e-06,
      "loss": 0.3587,
      "step": 8
    },
    {
      "epoch": 0.225,
      "grad_norm": 26.633220672607422,
      "learning_rate": 1e-06,
      "loss": 0.3009,
      "step": 9
    },
    {
      "epoch": 0.25,
      "grad_norm": 19.493844985961914,
      "learning_rate": 1e-06,
      "loss": 0.091,
      "step": 10
    },
    {
      "epoch": 0.275,
      "grad_norm": 9.516935348510742,
      "learning_rate": 1e-06,
      "loss": 0.045,
      "step": 11
    },
    {
      "epoch": 0.3,
      "grad_norm": 5.694634914398193,
      "learning_rate": 1e-06,
      "loss": 0.0358,
      "step": 12
    },
    {
      "epoch": 0.325,
      "grad_norm": 3.269968271255493,
      "learning_rate": 1e-06,
      "loss": 0.0218,
      "step": 13
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.6145660877227783,
      "learning_rate": 1e-06,
      "loss": 0.0161,
      "step": 14
    },
    {
      "epoch": 0.375,
      "grad_norm": 2.1988866329193115,
      "learning_rate": 1e-06,
      "loss": 0.0141,
      "step": 15
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.8373610973358154,
      "learning_rate": 1e-06,
      "loss": 0.0061,
      "step": 16
    },
    {
      "epoch": 0.425,
      "grad_norm": 1.4379240274429321,
      "learning_rate": 1e-06,
      "loss": 0.0065,
      "step": 17
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.604111671447754,
      "learning_rate": 1e-06,
      "loss": 0.0062,
      "step": 18
    },
    {
      "epoch": 0.475,
      "grad_norm": 2.3964450359344482,
      "learning_rate": 1e-06,
      "loss": 0.0062,
      "step": 19
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.2232977151870728,
      "learning_rate": 1e-06,
      "loss": 0.0042,
      "step": 20
    },
    {
      "epoch": 0.525,
      "grad_norm": 2.4161126613616943,
      "learning_rate": 1e-06,
      "loss": 0.0044,
      "step": 21
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.5427428483963013,
      "learning_rate": 1e-06,
      "loss": 0.0031,
      "step": 22
    },
    {
      "epoch": 0.575,
      "grad_norm": 0.6795447468757629,
      "learning_rate": 1e-06,
      "loss": 0.0008,
      "step": 23
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4597806930541992,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "step": 24
    },
    {
      "epoch": 0.625,
      "grad_norm": 1.3745927810668945,
      "learning_rate": 1e-06,
      "loss": 0.004,
      "step": 25
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 13389019086848.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}