Prompt Format:

### Instruction: {question}

### Response: {response}

Use these options to load with bnb:

4-bit params: {'load_in_4bit': True, 'bnb_4bit_compute_dtype':
torch.float16, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': True}

Downloads last month
11
Safetensors
Model size
38.5B params
Tensor type
F16
F32
U8
Inference Providers NEW
This model isn't deployed by any Inference Provider. 馃檵 Ask for provider support