last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3883495145631068,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.038834951456310676,
      "grad_norm": 21.15166664123535,
      "learning_rate": 2e-05,
      "loss": 8.4251,
      "step": 1
    },
    {
      "epoch": 0.038834951456310676,
      "eval_loss": 9.114510536193848,
      "eval_runtime": 5.2453,
      "eval_samples_per_second": 2.097,
      "eval_steps_per_second": 1.144,
      "step": 1
    },
    {
      "epoch": 0.07766990291262135,
      "grad_norm": 18.931537628173828,
      "learning_rate": 4e-05,
      "loss": 8.4191,
      "step": 2
    },
    {
      "epoch": 0.11650485436893204,
      "grad_norm": 33.92222595214844,
      "learning_rate": 6e-05,
      "loss": 9.326,
      "step": 3
    },
    {
      "epoch": 0.11650485436893204,
      "eval_loss": 8.625128746032715,
      "eval_runtime": 5.2629,
      "eval_samples_per_second": 2.09,
      "eval_steps_per_second": 1.14,
      "step": 3
    },
    {
      "epoch": 0.1553398058252427,
      "grad_norm": 17.89618682861328,
      "learning_rate": 8e-05,
      "loss": 7.6277,
      "step": 4
    },
    {
      "epoch": 0.1941747572815534,
      "grad_norm": 22.238508224487305,
      "learning_rate": 0.0001,
      "loss": 6.5102,
      "step": 5
    },
    {
      "epoch": 0.23300970873786409,
      "grad_norm": 17.981592178344727,
      "learning_rate": 0.00012,
      "loss": 5.852,
      "step": 6
    },
    {
      "epoch": 0.23300970873786409,
      "eval_loss": 4.704180717468262,
      "eval_runtime": 5.1699,
      "eval_samples_per_second": 2.128,
      "eval_steps_per_second": 1.161,
      "step": 6
    },
    {
      "epoch": 0.27184466019417475,
      "grad_norm": 33.65167236328125,
      "learning_rate": 0.00014,
      "loss": 5.1897,
      "step": 7
    },
    {
      "epoch": 0.3106796116504854,
      "grad_norm": 19.27935218811035,
      "learning_rate": 0.00016,
      "loss": 3.3586,
      "step": 8
    },
    {
      "epoch": 0.34951456310679613,
      "grad_norm": 15.050045013427734,
      "learning_rate": 0.00018,
      "loss": 2.734,
      "step": 9
    },
    {
      "epoch": 0.34951456310679613,
      "eval_loss": 2.383453845977783,
      "eval_runtime": 5.0182,
      "eval_samples_per_second": 2.192,
      "eval_steps_per_second": 1.196,
      "step": 9
    },
    {
      "epoch": 0.3883495145631068,
      "grad_norm": 31.062732696533203,
      "learning_rate": 0.0002,
      "loss": 2.3454,
      "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": 1.483774567120896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}