highex_l9-17_lowag_l18-26
Merged LoRA adapter produced for the EMNLP ablation study.
LoRA configuration
- Base model:
unsloth/qwen3-8b-unsloth-bnb-4bit - r: 64
- lora_alpha: 128
- lora_dropout: 0.0
- target_modules: ['up_proj', 'down_proj', 'o_proj', 'gate_proj', 'v_proj', 'q_proj', 'k_proj']
- layers_to_transform: [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26]
- Downloads last month
- 1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support