llama2_7b_responsibility-16R-128B-4E-100S/training_log.json

{
  "base_model_name": "meta-llama_Llama-2-7b-chat-hf",
  "base_model_class": "LlamaForCausalLM",
  "base_loaded_in_4bit": false,
  "base_loaded_in_8bit": false,
  "projections": "q, v, k, o",
  "eval_loss": 0.2940330505371094,
  "eval_runtime": 178.282,
  "eval_samples_per_second": 8.666,
  "eval_steps_per_second": 1.088,
  "epoch": 4.0,
  "current_steps": 4701,
  "loss": 0.2961,
  "learning_rate": 4.195804195804196e-06,
  "train_runtime": 10950.2359,
  "train_samples_per_second": 1.719,
  "train_steps_per_second": 0.013,
  "total_flos": 5.982058047995904e+17,
  "train_loss": 0.4049693025699278
}