training_log.json

{
  "base_model_name": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
  "base_model_class": "LlamaForCausalLM",
  "base_loaded_in_4bit": false,
  "base_loaded_in_8bit": true,
  "projections": "q, k, v, o",
  "loss": 1.4473,
  "learning_rate": 2.364532019704433e-06,
  "epoch": 0.25,
  "current_steps": 2149,
  "current_steps_adjusted": 2149,
  "epoch_adjusted": 0.25,
  "train_runtime": 6758.723,
  "train_samples_per_second": 1.907,
  "train_steps_per_second": 0.159,
  "total_flos": 6.25660356722688e+16,
  "train_loss": 1.498897494826206
}