checkpoint-17396/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 17396,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00022993791676247414,
      "grad_norm": 11.149688720703125,
      "learning_rate": 0.0,
      "loss": 5.3185,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 13.886802673339844,
      "learning_rate": 3.124640598044854e-05,
      "loss": 2.6529,
      "step": 4349
    },
    {
      "epoch": 1.0,
      "eval_cosine_accuracy": 0.9491846561431885,
      "eval_loss": 1.4502098560333252,
      "eval_runtime": 24.744,
      "eval_samples_per_second": 384.133,
      "eval_steps_per_second": 3.031,
      "step": 4349
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.237020015716553,
      "learning_rate": 3.574125759317025e-05,
      "loss": 1.8024,
      "step": 8698
    },
    {
      "epoch": 2.0,
      "eval_cosine_accuracy": 0.9600210189819336,
      "eval_loss": 1.399303674697876,
      "eval_runtime": 26.5029,
      "eval_samples_per_second": 358.64,
      "eval_steps_per_second": 2.83,
      "step": 8698
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.765737533569336,
      "learning_rate": 5.7461828927926447e-08,
      "loss": 1.4655,
      "step": 13047
    },
    {
      "epoch": 3.0,
      "eval_cosine_accuracy": 0.9607574939727783,
      "eval_loss": 1.3219194412231445,
      "eval_runtime": 24.731,
      "eval_samples_per_second": 384.335,
      "eval_steps_per_second": 3.033,
      "step": 13047
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.971332550048828,
      "learning_rate": 2.274532344027589e-05,
      "loss": 1.3564,
      "step": 17396
    },
    {
      "epoch": 4.0,
      "eval_cosine_accuracy": 0.9600210189819336,
      "eval_loss": 1.3028672933578491,
      "eval_runtime": 26.3111,
      "eval_samples_per_second": 361.254,
      "eval_steps_per_second": 2.851,
      "step": 17396
    }
  ],
  "logging_steps": 500,
  "max_steps": 26094,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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
}