Downloads · 30 days
5
5% of all-time downloads
kz-history-qa-llama3.1-8b
timkaiyr/kz-history-qa-llama3.1-8b
kz-history-qa-llama3.1-8b is a question answering model from timkaiyr. Use it when the input is a question plus a passage. It is set up for peft. The card lists the license as llama3.1.
QLoRA adapter for unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit, fine-tuned to answer questions about the history of Kazakhstan (11th-grade school textbook) in Russian.
All-time downloads
100
Public
Repo size
521 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors168 MB · 91%
At a glance
- Task
- Question Answering
- Library
- peft
- License
- llama3.1
- Access
- Public
- Created
- Jul 1, 2026
- Updated
- Jul 5, 2026
- SHA
- bf17ca7b
Try a prompt
Base models
Model card
From the Hugging Face model README
kz-history-qa-llama3.1-8b
QLoRA adapter for unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit, fine-tuned to answer questions
about the history of Kazakhstan (11th-grade school textbook) in Russian.
Data
- Source: OCR of the textbook pages.
- Dataset: 367 question-answer pairs —
timkaiyr/kz-history-qa-dataset.
Training
- Method: SFT + QLoRA (Unsloth), 4-bit.
- Epochs: 3; learning rate: 0.0002; max_length: 1024.
- LoRA: r=16, alpha=16, dropout=0.
Usage
Load the base model in 4-bit and apply the adapter via PEFT.
- Task
- Question Answering
- Library
- peft
- License
- llama3.1
- Languages
- ru
- Created
- Jul 1, 2026
- Updated
- Jul 5, 2026