last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03262642740619902,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0032626427406199023,
      "grad_norm": 0.7658309936523438,
      "learning_rate": 2e-05,
      "loss": 1.7421,
      "step": 1
    },
    {
      "epoch": 0.0032626427406199023,
      "eval_loss": 1.6826673746109009,
      "eval_runtime": 3.8278,
      "eval_samples_per_second": 33.962,
      "eval_steps_per_second": 16.981,
      "step": 1
    },
    {
      "epoch": 0.0065252854812398045,
      "grad_norm": 0.6731956005096436,
      "learning_rate": 4e-05,
      "loss": 1.3735,
      "step": 2
    },
    {
      "epoch": 0.009787928221859706,
      "grad_norm": 0.8331893086433411,
      "learning_rate": 6e-05,
      "loss": 1.6503,
      "step": 3
    },
    {
      "epoch": 0.009787928221859706,
      "eval_loss": 1.6817997694015503,
      "eval_runtime": 2.9268,
      "eval_samples_per_second": 44.417,
      "eval_steps_per_second": 22.208,
      "step": 3
    },
    {
      "epoch": 0.013050570962479609,
      "grad_norm": 0.6314859986305237,
      "learning_rate": 8e-05,
      "loss": 1.3417,
      "step": 4
    },
    {
      "epoch": 0.01631321370309951,
      "grad_norm": 0.7163793444633484,
      "learning_rate": 0.0001,
      "loss": 1.5046,
      "step": 5
    },
    {
      "epoch": 0.01957585644371941,
      "grad_norm": 0.614728569984436,
      "learning_rate": 0.00012,
      "loss": 1.4456,
      "step": 6
    },
    {
      "epoch": 0.01957585644371941,
      "eval_loss": 1.6709381341934204,
      "eval_runtime": 2.8447,
      "eval_samples_per_second": 45.699,
      "eval_steps_per_second": 22.849,
      "step": 6
    },
    {
      "epoch": 0.022838499184339316,
      "grad_norm": 0.8054418563842773,
      "learning_rate": 0.00014,
      "loss": 1.8589,
      "step": 7
    },
    {
      "epoch": 0.026101141924959218,
      "grad_norm": 0.853880763053894,
      "learning_rate": 0.00016,
      "loss": 2.0009,
      "step": 8
    },
    {
      "epoch": 0.02936378466557912,
      "grad_norm": 0.7069666981697083,
      "learning_rate": 0.00018,
      "loss": 1.282,
      "step": 9
    },
    {
      "epoch": 0.02936378466557912,
      "eval_loss": 1.6430575847625732,
      "eval_runtime": 2.8832,
      "eval_samples_per_second": 45.089,
      "eval_steps_per_second": 22.544,
      "step": 9
    },
    {
      "epoch": 0.03262642740619902,
      "grad_norm": 0.6495545506477356,
      "learning_rate": 0.0002,
      "loss": 1.4099,
      "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": 109331418710016.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}