| { | |
| "base_model_name": "Llama-2-13b-hf", | |
| "base_model_class": "LlamaForCausalLM", | |
| "base_loaded_in_4bit": true, | |
| "base_loaded_in_8bit": false, | |
| "projections": "q, v", | |
| "loss": 2.1927, | |
| "grad_norm": 0.9019927978515625, | |
| "learning_rate": 2.5456088247772588e-09, | |
| "epoch": 3.0, | |
| "current_steps": 2456, | |
| "current_steps_adjusted": 2456, | |
| "epoch_adjusted": 3.0, | |
| "train_runtime": 6265.7426, | |
| "train_samples_per_second": 1.568, | |
| "train_steps_per_second": 0.392, | |
| "total_flos": 1.94348076171264e+17, | |
| "train_loss": 2.3536050414282177 | |
| } |