{ "repo_id": "broadfield-dev/savant_2_gsm8k_final", "resume_from_benchmark": "", "benchmark_name_prefix": "stable-run", "benchmark_interval": 20, "sft_epochs": 4, "sft_lr": 2e-05, "sft_warmup_proportion": 0.1, "rl_lr": 1.5e-06, "sft_slice": "0:2000", "rl_slice": "4000:7473", "context_length": 1024, "base_model_name": "distilgpt2", "use_scheduler": true, "sft_batch_size": 1, "grad_accum_steps": 16 }