last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00048693196343140954,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.6231065447713652e-05,
      "eval_loss": 1.163723349571228,
      "eval_runtime": 4571.4981,
      "eval_samples_per_second": 5.675,
      "eval_steps_per_second": 2.837,
      "step": 1
    },
    {
      "epoch": 4.8693196343140954e-05,
      "grad_norm": 0.12008862942457199,
      "learning_rate": 6e-05,
      "loss": 0.7991,
      "step": 3
    },
    {
      "epoch": 9.738639268628191e-05,
      "grad_norm": 0.1235838234424591,
      "learning_rate": 0.00012,
      "loss": 0.8886,
      "step": 6
    },
    {
      "epoch": 0.00012984852358170922,
      "eval_loss": 1.160834550857544,
      "eval_runtime": 4586.6402,
      "eval_samples_per_second": 5.656,
      "eval_steps_per_second": 2.828,
      "step": 8
    },
    {
      "epoch": 0.00014607958902942286,
      "grad_norm": 0.16633595526218414,
      "learning_rate": 0.00018,
      "loss": 1.1805,
      "step": 9
    },
    {
      "epoch": 0.00019477278537256382,
      "grad_norm": 0.19877596199512482,
      "learning_rate": 0.00019510565162951537,
      "loss": 1.0945,
      "step": 12
    },
    {
      "epoch": 0.00024346598171570477,
      "grad_norm": 0.26173800230026245,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.8844,
      "step": 15
    },
    {
      "epoch": 0.00025969704716341844,
      "eval_loss": 1.137023687362671,
      "eval_runtime": 4585.1131,
      "eval_samples_per_second": 5.658,
      "eval_steps_per_second": 2.829,
      "step": 16
    },
    {
      "epoch": 0.0002921591780588457,
      "grad_norm": 0.21635204553604126,
      "learning_rate": 0.00013090169943749476,
      "loss": 1.0719,
      "step": 18
    },
    {
      "epoch": 0.0003408523744019867,
      "grad_norm": 0.27578991651535034,
      "learning_rate": 8.435655349597689e-05,
      "loss": 1.0951,
      "step": 21
    },
    {
      "epoch": 0.00038954557074512763,
      "grad_norm": 0.3835023045539856,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.9924,
      "step": 24
    },
    {
      "epoch": 0.00038954557074512763,
      "eval_loss": 1.119747281074524,
      "eval_runtime": 4586.0874,
      "eval_samples_per_second": 5.657,
      "eval_steps_per_second": 2.828,
      "step": 24
    },
    {
      "epoch": 0.0004382387670882686,
      "grad_norm": 0.18901470303535461,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.9938,
      "step": 27
    },
    {
      "epoch": 0.00048693196343140954,
      "grad_norm": 0.2615954279899597,
      "learning_rate": 0.0,
      "loss": 0.9401,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 70,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6355377153638400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}