last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0051013901288101005,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005101390128810101,
      "grad_norm": 4.33687686920166,
      "learning_rate": 2e-05,
      "loss": 3.4878,
      "step": 1
    },
    {
      "epoch": 0.0005101390128810101,
      "eval_loss": 0.9874624013900757,
      "eval_runtime": 42.3405,
      "eval_samples_per_second": 19.508,
      "eval_steps_per_second": 9.754,
      "step": 1
    },
    {
      "epoch": 0.0010202780257620202,
      "grad_norm": 4.788012981414795,
      "learning_rate": 4e-05,
      "loss": 3.7314,
      "step": 2
    },
    {
      "epoch": 0.0015304170386430302,
      "grad_norm": 7.3934454917907715,
      "learning_rate": 6e-05,
      "loss": 4.4561,
      "step": 3
    },
    {
      "epoch": 0.0015304170386430302,
      "eval_loss": 0.9790695905685425,
      "eval_runtime": 40.9953,
      "eval_samples_per_second": 20.149,
      "eval_steps_per_second": 10.074,
      "step": 3
    },
    {
      "epoch": 0.0020405560515240405,
      "grad_norm": 4.933080196380615,
      "learning_rate": 8e-05,
      "loss": 3.8327,
      "step": 4
    },
    {
      "epoch": 0.0025506950644050503,
      "grad_norm": 4.43290376663208,
      "learning_rate": 0.0001,
      "loss": 3.3751,
      "step": 5
    },
    {
      "epoch": 0.0030608340772860605,
      "grad_norm": 4.0239644050598145,
      "learning_rate": 0.00012,
      "loss": 3.7749,
      "step": 6
    },
    {
      "epoch": 0.0030608340772860605,
      "eval_loss": 0.9542436003684998,
      "eval_runtime": 41.0527,
      "eval_samples_per_second": 20.12,
      "eval_steps_per_second": 10.06,
      "step": 6
    },
    {
      "epoch": 0.0035709730901670707,
      "grad_norm": 3.282405376434326,
      "learning_rate": 0.00014,
      "loss": 4.1165,
      "step": 7
    },
    {
      "epoch": 0.004081112103048081,
      "grad_norm": 3.7907450199127197,
      "learning_rate": 0.00016,
      "loss": 4.5457,
      "step": 8
    },
    {
      "epoch": 0.00459125111592909,
      "grad_norm": 3.6256449222564697,
      "learning_rate": 0.00018,
      "loss": 3.8068,
      "step": 9
    },
    {
      "epoch": 0.00459125111592909,
      "eval_loss": 0.9475084543228149,
      "eval_runtime": 41.0737,
      "eval_samples_per_second": 20.11,
      "eval_steps_per_second": 10.055,
      "step": 9
    },
    {
      "epoch": 0.0051013901288101005,
      "grad_norm": 4.504990100860596,
      "learning_rate": 0.0002,
      "loss": 3.0924,
      "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": 3330076418506752.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}