last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0049800796812749,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00049800796812749,
      "grad_norm": 1.4692789316177368,
      "learning_rate": 2e-05,
      "loss": 2.8063,
      "step": 1
    },
    {
      "epoch": 0.00049800796812749,
      "eval_loss": 2.97845196723938,
      "eval_runtime": 45.5862,
      "eval_samples_per_second": 18.558,
      "eval_steps_per_second": 9.279,
      "step": 1
    },
    {
      "epoch": 0.00099601593625498,
      "grad_norm": 1.4389501810073853,
      "learning_rate": 4e-05,
      "loss": 2.8635,
      "step": 2
    },
    {
      "epoch": 0.00149402390438247,
      "grad_norm": 2.2483975887298584,
      "learning_rate": 6e-05,
      "loss": 3.4006,
      "step": 3
    },
    {
      "epoch": 0.00149402390438247,
      "eval_loss": 2.9519526958465576,
      "eval_runtime": 45.5115,
      "eval_samples_per_second": 18.589,
      "eval_steps_per_second": 9.294,
      "step": 3
    },
    {
      "epoch": 0.00199203187250996,
      "grad_norm": 1.5795270204544067,
      "learning_rate": 8e-05,
      "loss": 2.7943,
      "step": 4
    },
    {
      "epoch": 0.00249003984063745,
      "grad_norm": 1.3678991794586182,
      "learning_rate": 0.0001,
      "loss": 2.6807,
      "step": 5
    },
    {
      "epoch": 0.00298804780876494,
      "grad_norm": 1.5033429861068726,
      "learning_rate": 0.00012,
      "loss": 2.8275,
      "step": 6
    },
    {
      "epoch": 0.00298804780876494,
      "eval_loss": 2.6760687828063965,
      "eval_runtime": 45.6888,
      "eval_samples_per_second": 18.517,
      "eval_steps_per_second": 9.258,
      "step": 6
    },
    {
      "epoch": 0.0034860557768924302,
      "grad_norm": 1.0134755373001099,
      "learning_rate": 0.00014,
      "loss": 2.4683,
      "step": 7
    },
    {
      "epoch": 0.00398406374501992,
      "grad_norm": 0.830837607383728,
      "learning_rate": 0.00016,
      "loss": 2.5534,
      "step": 8
    },
    {
      "epoch": 0.004482071713147411,
      "grad_norm": 0.8642377853393555,
      "learning_rate": 0.00018,
      "loss": 2.5087,
      "step": 9
    },
    {
      "epoch": 0.004482071713147411,
      "eval_loss": 2.5361790657043457,
      "eval_runtime": 45.6708,
      "eval_samples_per_second": 18.524,
      "eval_steps_per_second": 9.262,
      "step": 9
    },
    {
      "epoch": 0.0049800796812749,
      "grad_norm": 0.9797606468200684,
      "learning_rate": 0.0002,
      "loss": 2.4235,
      "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": 2514182181224448.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}