checkpoint-4733/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 4733,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00021128248468201986,
      "grad_norm": 7.594412803649902,
      "learning_rate": 0.0,
      "loss": 2.6278,
      "step": 1
    },
    {
      "epoch": 0.21128248468201985,
      "grad_norm": 6.050217628479004,
      "learning_rate": 8.441064638783271e-06,
      "loss": 2.7281,
      "step": 1000
    },
    {
      "epoch": 0.21128248468201985,
      "eval_cosine_accuracy": 0.9345883131027222,
      "eval_loss": 0.6125919222831726,
      "eval_runtime": 34.2392,
      "eval_samples_per_second": 277.723,
      "eval_steps_per_second": 2.19,
      "step": 1000
    },
    {
      "epoch": 0.4225649693640397,
      "grad_norm": 6.340444087982178,
      "learning_rate": 1.6890578791719477e-05,
      "loss": 1.999,
      "step": 2000
    },
    {
      "epoch": 0.4225649693640397,
      "eval_cosine_accuracy": 0.9388999938964844,
      "eval_loss": 0.5676162838935852,
      "eval_runtime": 34.1182,
      "eval_samples_per_second": 278.708,
      "eval_steps_per_second": 2.198,
      "step": 2000
    },
    {
      "epoch": 0.6338474540460596,
      "grad_norm": 8.218648910522461,
      "learning_rate": 1.940745609916424e-05,
      "loss": 1.5073,
      "step": 3000
    },
    {
      "epoch": 0.6338474540460596,
      "eval_cosine_accuracy": 0.9328005313873291,
      "eval_loss": 0.5519113540649414,
      "eval_runtime": 34.1651,
      "eval_samples_per_second": 278.325,
      "eval_steps_per_second": 2.195,
      "step": 3000
    },
    {
      "epoch": 0.8451299387280794,
      "grad_norm": 9.030098915100098,
      "learning_rate": 1.846933984411682e-05,
      "loss": 0.9687,
      "step": 4000
    },
    {
      "epoch": 0.8451299387280794,
      "eval_cosine_accuracy": 0.9312230348587036,
      "eval_loss": 0.5424668788909912,
      "eval_runtime": 34.165,
      "eval_samples_per_second": 278.326,
      "eval_steps_per_second": 2.195,
      "step": 4000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 23665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}