casehold-lora / adapter_config.json
epochzero-0's picture
Adapter trained on 45,000 CaseHOLD questions, 85.92% on test
fd5a7ee verified
Raw History Blame Contribute Delete
181 Bytes
{
"base_model": "Qwen/Qwen2.5-0.5B",
"r": 16,
"alpha": 32,
"targets": [
"q_proj",
"k_proj",
"v_proj",
"o_proj"
],
"max_len": 640,
"trained_on": 45000
}