trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 14.16061019897461,
      "learning_rate": 0.0004,
      "loss": 3.5313,
      "step": 1
    },
    {
      "epoch": 0.1111111111111111,
      "eval_loss": 3.7931530475616455,
      "eval_runtime": 9.5456,
      "eval_samples_per_second": 7.438,
      "eval_steps_per_second": 0.524,
      "step": 1
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 13.478278160095215,
      "learning_rate": 0.00035,
      "loss": 3.4173,
      "step": 2
    },
    {
      "epoch": 0.2222222222222222,
      "eval_loss": 1.4140301942825317,
      "eval_runtime": 9.4126,
      "eval_samples_per_second": 7.543,
      "eval_steps_per_second": 0.531,
      "step": 2
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 2.5410170555114746,
      "learning_rate": 0.00030000000000000003,
      "loss": 1.1984,
      "step": 3
    },
    {
      "epoch": 0.3333333333333333,
      "eval_loss": 1.2153092622756958,
      "eval_runtime": 9.4229,
      "eval_samples_per_second": 7.535,
      "eval_steps_per_second": 0.531,
      "step": 3
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.9833934307098389,
      "learning_rate": 0.00025,
      "loss": 0.9172,
      "step": 4
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.2972338199615479,
      "eval_runtime": 9.3699,
      "eval_samples_per_second": 7.577,
      "eval_steps_per_second": 0.534,
      "step": 4
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 1.382117748260498,
      "learning_rate": 0.0002,
      "loss": 0.8168,
      "step": 5
    },
    {
      "epoch": 0.5555555555555556,
      "eval_loss": 1.2850072383880615,
      "eval_runtime": 9.4878,
      "eval_samples_per_second": 7.483,
      "eval_steps_per_second": 0.527,
      "step": 5
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.6453934907913208,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.7473,
      "step": 6
    },
    {
      "epoch": 0.6666666666666666,
      "eval_loss": 1.2965147495269775,
      "eval_runtime": 9.4605,
      "eval_samples_per_second": 7.505,
      "eval_steps_per_second": 0.529,
      "step": 6
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 0.432075172662735,
      "learning_rate": 0.0001,
      "loss": 0.701,
      "step": 7
    },
    {
      "epoch": 0.7777777777777778,
      "eval_loss": 1.3253140449523926,
      "eval_runtime": 9.4624,
      "eval_samples_per_second": 7.503,
      "eval_steps_per_second": 0.528,
      "step": 7
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.38340702652931213,
      "learning_rate": 5e-05,
      "loss": 0.676,
      "step": 8
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.339978575706482,
      "eval_runtime": 9.5112,
      "eval_samples_per_second": 7.465,
      "eval_steps_per_second": 0.526,
      "step": 8
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.3342841565608978,
      "learning_rate": 0.0,
      "loss": 0.6765,
      "step": 9
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.3411067724227905,
      "eval_runtime": 9.4591,
      "eval_samples_per_second": 7.506,
      "eval_steps_per_second": 0.529,
      "step": 9
    },
    {
      "epoch": 1.0,
      "step": 9,
      "total_flos": 6367416057593856.0,
      "train_loss": 1.409088393052419,
      "train_runtime": 210.617,
      "train_samples_per_second": 1.301,
      "train_steps_per_second": 0.043
    }
  ],
  "logging_steps": 1,
  "max_steps": 9,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6367416057593856.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}