Downloads · 30 days
0
task6
truettwo/task6
task6 is a question answering model from truettwo. Use it when the input is a question plus a passage. The card lists the license as llama3.2.
- r=16, alpha=32, dropout=0.05 - maxsteps=500, lr=2e-4, warmup=50 - 4-bit NF4 QLoRA (unsloth)
Access
Public
Updated May 17, 2026
Repo size
62.3 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors45.1 MB · 72%
At a glance
- Task
- Question Answering
- License
- llama3.2
- Access
- Public
- Created
- Apr 30, 2026
- Updated
- May 17, 2026
- SHA
- bddfd97a
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
Llama-3.2-1B + LoRA — SQuAD 2.0 QA
Результаты
| Метод | F1 | Exact Match |
|---|---|---|
| Zero-shot | 8.28% | 4.0% |
| Few-shot (5) | 0.0% | 0.0% |
| LoRA fine-tuned | 61.63% | 43.0% |
Конфигурация LoRA
r=16,alpha=32,dropout=0.05max_steps=500,lr=2e-4,warmup=50- 4-bit NF4 QLoRA (unsloth)
- Task
- Question Answering
- License
- llama3.2
- Languages
- en
- Created
- Apr 30, 2026
- Updated
- May 17, 2026