RLVR_VFINAL_STEP30 / prime_lora_adapter_export.json

RLVR dense-reward LoRA adapter step 30

89ca56e verified 4 months ago

216 Bytes

{
  "base_model": "JayZenith/SFT_HALF_A_V8",
  "adapter_dir": "outputs/RLVR_SIGNAL_V4002_POOL_B_DENSE_LORA_R64_A128/run_default/broadcasts/step_30",
  "rank": 64,
  "alpha": 128.0,
  "lora_a": 252,
  "lora_b": 252
}

	{
	"base_model": "JayZenith/SFT_HALF_A_V8",
	"adapter_dir": "outputs/RLVR_SIGNAL_V4002_POOL_B_DENSE_LORA_R64_A128/run_default/broadcasts/step_30",
	"rank": 64,
	"alpha": 128.0,
	"lora_a": 252,
	"lora_b": 252
	}