llama2_13b_responsibility-16R-128B-4E-100S/training_log.json

{
  "base_model_name": "meta-llama_Llama-2-13b-chat-hf",
  "base_model_class": "LlamaForCausalLM",
  "base_loaded_in_4bit": false,
  "base_loaded_in_8bit": false,
  "projections": "q, v, k, o",
  "eval_loss": 0.27035561203956604,
  "eval_runtime": 318.5835,
  "eval_samples_per_second": 4.85,
  "eval_steps_per_second": 0.609,
  "epoch": 4.0,
  "current_steps": 6268,
  "loss": 0.262,
  "learning_rate": 1.5624999999999999e-06,
  "train_runtime": 27128.8237,
  "train_samples_per_second": 0.925,
  "train_steps_per_second": 0.007,
  "total_flos": 1.9848229987811328e+18,
  "train_loss": 0.34869462008378943
}