last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001594896331738437,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001594896331738437,
      "grad_norm": 1.502610206604004,
      "learning_rate": 2e-05,
      "loss": 5.1059,
      "step": 1
    },
    {
      "epoch": 0.0001594896331738437,
      "eval_loss": 1.1597040891647339,
      "eval_runtime": 149.4566,
      "eval_samples_per_second": 17.664,
      "eval_steps_per_second": 8.832,
      "step": 1
    },
    {
      "epoch": 0.0003189792663476874,
      "grad_norm": 1.2862768173217773,
      "learning_rate": 4e-05,
      "loss": 4.217,
      "step": 2
    },
    {
      "epoch": 0.0004784688995215311,
      "grad_norm": 1.669655203819275,
      "learning_rate": 6e-05,
      "loss": 5.4411,
      "step": 3
    },
    {
      "epoch": 0.0004784688995215311,
      "eval_loss": 1.1572209596633911,
      "eval_runtime": 149.2613,
      "eval_samples_per_second": 17.687,
      "eval_steps_per_second": 8.844,
      "step": 3
    },
    {
      "epoch": 0.0006379585326953748,
      "grad_norm": 1.7106208801269531,
      "learning_rate": 8e-05,
      "loss": 4.1717,
      "step": 4
    },
    {
      "epoch": 0.0007974481658692185,
      "grad_norm": 1.4486793279647827,
      "learning_rate": 0.0001,
      "loss": 4.3287,
      "step": 5
    },
    {
      "epoch": 0.0009569377990430622,
      "grad_norm": 1.6041182279586792,
      "learning_rate": 0.00012,
      "loss": 4.6163,
      "step": 6
    },
    {
      "epoch": 0.0009569377990430622,
      "eval_loss": 1.1258790493011475,
      "eval_runtime": 149.8099,
      "eval_samples_per_second": 17.622,
      "eval_steps_per_second": 8.811,
      "step": 6
    },
    {
      "epoch": 0.0011164274322169058,
      "grad_norm": 2.0097925662994385,
      "learning_rate": 0.00014,
      "loss": 4.1186,
      "step": 7
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": 2.1436350345611572,
      "learning_rate": 0.00016,
      "loss": 4.1567,
      "step": 8
    },
    {
      "epoch": 0.0014354066985645933,
      "grad_norm": 1.7882678508758545,
      "learning_rate": 0.00018,
      "loss": 3.1176,
      "step": 9
    },
    {
      "epoch": 0.0014354066985645933,
      "eval_loss": 1.034005045890808,
      "eval_runtime": 149.7222,
      "eval_samples_per_second": 17.633,
      "eval_steps_per_second": 8.816,
      "step": 9
    },
    {
      "epoch": 0.001594896331738437,
      "grad_norm": 1.6461585760116577,
      "learning_rate": 0.0002,
      "loss": 4.1039,
      "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": 3403669485649920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}