checkpoint-5057/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 5057,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00019774569903104609,
      "grad_norm": 5.842831611633301,
      "learning_rate": 0.0,
      "loss": 0.8464,
      "step": 1
    },
    {
      "epoch": 0.19774569903104608,
      "grad_norm": 6.009142875671387,
      "learning_rate": 3.290513833992095e-05,
      "loss": 0.8746,
      "step": 1000
    },
    {
      "epoch": 0.19774569903104608,
      "eval_cosine_accuracy": 0.9700284004211426,
      "eval_loss": 0.40132495760917664,
      "eval_runtime": 35.9259,
      "eval_samples_per_second": 264.684,
      "eval_steps_per_second": 2.088,
      "step": 1000
    },
    {
      "epoch": 0.39549139806209216,
      "grad_norm": 5.525256156921387,
      "learning_rate": 4.823848238482385e-05,
      "loss": 0.8834,
      "step": 2000
    },
    {
      "epoch": 0.39549139806209216,
      "eval_cosine_accuracy": 0.9702387452125549,
      "eval_loss": 0.4074099063873291,
      "eval_runtime": 34.7469,
      "eval_samples_per_second": 273.665,
      "eval_steps_per_second": 2.158,
      "step": 2000
    },
    {
      "epoch": 0.5932370970931382,
      "grad_norm": 6.155282497406006,
      "learning_rate": 4.4576283600673844e-05,
      "loss": 0.7973,
      "step": 3000
    },
    {
      "epoch": 0.5932370970931382,
      "eval_cosine_accuracy": 0.9673992991447449,
      "eval_loss": 0.41063427925109863,
      "eval_runtime": 34.6206,
      "eval_samples_per_second": 274.663,
      "eval_steps_per_second": 2.166,
      "step": 3000
    },
    {
      "epoch": 0.7909827961241843,
      "grad_norm": 7.556669235229492,
      "learning_rate": 4.0917747015308e-05,
      "loss": 0.5365,
      "step": 4000
    },
    {
      "epoch": 0.7909827961241843,
      "eval_cosine_accuracy": 0.9680302739143372,
      "eval_loss": 0.3832950294017792,
      "eval_runtime": 34.6136,
      "eval_samples_per_second": 274.718,
      "eval_steps_per_second": 2.167,
      "step": 4000
    },
    {
      "epoch": 0.9887284951552304,
      "grad_norm": 11.179519653320312,
      "learning_rate": 3.725554823115799e-05,
      "loss": 0.4558,
      "step": 5000
    },
    {
      "epoch": 0.9887284951552304,
      "eval_cosine_accuracy": 0.9672941565513611,
      "eval_loss": 0.37456464767456055,
      "eval_runtime": 34.7639,
      "eval_samples_per_second": 273.531,
      "eval_steps_per_second": 2.157,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 15171,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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
}