last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4521193092621664,
  "eval_steps": 9,
  "global_step": 27,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01674515960230246,
      "eval_loss": 1.4724388122558594,
      "eval_runtime": 73.0963,
      "eval_samples_per_second": 2.763,
      "eval_steps_per_second": 1.382,
      "step": 1
    },
    {
      "epoch": 0.05023547880690738,
      "grad_norm": 102.90454864501953,
      "learning_rate": 3e-05,
      "loss": 45.0108,
      "step": 3
    },
    {
      "epoch": 0.10047095761381476,
      "grad_norm": 66.23411560058594,
      "learning_rate": 6e-05,
      "loss": 42.1127,
      "step": 6
    },
    {
      "epoch": 0.15070643642072212,
      "grad_norm": 49.03892517089844,
      "learning_rate": 9e-05,
      "loss": 25.533,
      "step": 9
    },
    {
      "epoch": 0.15070643642072212,
      "eval_loss": 0.48065412044525146,
      "eval_runtime": 73.1001,
      "eval_samples_per_second": 2.763,
      "eval_steps_per_second": 1.382,
      "step": 9
    },
    {
      "epoch": 0.20094191522762953,
      "grad_norm": 40.06981658935547,
      "learning_rate": 0.00012,
      "loss": 12.9041,
      "step": 12
    },
    {
      "epoch": 0.25117739403453687,
      "grad_norm": 18.526214599609375,
      "learning_rate": 0.00015000000000000001,
      "loss": 8.4587,
      "step": 15
    },
    {
      "epoch": 0.30141287284144425,
      "grad_norm": 26.23703384399414,
      "learning_rate": 0.00018,
      "loss": 6.8537,
      "step": 18
    },
    {
      "epoch": 0.30141287284144425,
      "eval_loss": 0.16974753141403198,
      "eval_runtime": 73.0714,
      "eval_samples_per_second": 2.764,
      "eval_steps_per_second": 1.382,
      "step": 18
    },
    {
      "epoch": 0.3516483516483517,
      "grad_norm": 25.50685691833496,
      "learning_rate": 0.0001999229036240723,
      "loss": 5.2731,
      "step": 21
    },
    {
      "epoch": 0.40188383045525905,
      "grad_norm": 16.794132232666016,
      "learning_rate": 0.00019876883405951377,
      "loss": 3.9976,
      "step": 24
    },
    {
      "epoch": 0.4521193092621664,
      "grad_norm": 16.295429229736328,
      "learning_rate": 0.00019624552364536473,
      "loss": 3.9352,
      "step": 27
    },
    {
      "epoch": 0.4521193092621664,
      "eval_loss": 0.1071891188621521,
      "eval_runtime": 73.2333,
      "eval_samples_per_second": 2.758,
      "eval_steps_per_second": 1.379,
      "step": 27
    }
  ],
  "logging_steps": 3,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 9,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.419958289474519e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}