last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.10666666666666667,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035555555555555557,
      "eval_loss": 4.701450824737549,
      "eval_runtime": 7.1119,
      "eval_samples_per_second": 16.732,
      "eval_steps_per_second": 8.437,
      "step": 1
    },
    {
      "epoch": 0.010666666666666666,
      "grad_norm": 8.527983665466309,
      "learning_rate": 6e-05,
      "loss": 4.5862,
      "step": 3
    },
    {
      "epoch": 0.021333333333333333,
      "grad_norm": 6.9240522384643555,
      "learning_rate": 0.00012,
      "loss": 3.9541,
      "step": 6
    },
    {
      "epoch": 0.028444444444444446,
      "eval_loss": 1.967842698097229,
      "eval_runtime": 7.0506,
      "eval_samples_per_second": 16.878,
      "eval_steps_per_second": 8.51,
      "step": 8
    },
    {
      "epoch": 0.032,
      "grad_norm": 4.608461856842041,
      "learning_rate": 0.00018,
      "loss": 2.5516,
      "step": 9
    },
    {
      "epoch": 0.042666666666666665,
      "grad_norm": 4.794651985168457,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.9364,
      "step": 12
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 2.363208055496216,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.2801,
      "step": 15
    },
    {
      "epoch": 0.05688888888888889,
      "eval_loss": 0.507258415222168,
      "eval_runtime": 7.0775,
      "eval_samples_per_second": 16.814,
      "eval_steps_per_second": 8.478,
      "step": 16
    },
    {
      "epoch": 0.064,
      "grad_norm": 5.319071292877197,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.4827,
      "step": 18
    },
    {
      "epoch": 0.07466666666666667,
      "grad_norm": 1.236522912979126,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.2749,
      "step": 21
    },
    {
      "epoch": 0.08533333333333333,
      "grad_norm": 0.825536847114563,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.27,
      "step": 24
    },
    {
      "epoch": 0.08533333333333333,
      "eval_loss": 0.20560859143733978,
      "eval_runtime": 7.1295,
      "eval_samples_per_second": 16.691,
      "eval_steps_per_second": 8.416,
      "step": 24
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.8618375658988953,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.2052,
      "step": 27
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 0.8433903455734253,
      "learning_rate": 0.0,
      "loss": 0.193,
      "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": 4939038012211200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}