RichardErkhov/haripritam_-_Qwen2-0.5B-fncl-2-gguf

Quantization made by Richard Erkhov.

Qwen2-0.5B-fncl-2 - GGUF

Model creator: https://huggingface.co/haripritam/
Original model: https://huggingface.co/haripritam/Qwen2-0.5B-fncl-2/

Name	Quant method	Size
Qwen2-0.5B-fncl-2.Q2_K.gguf	Q2_K	0.32GB
Qwen2-0.5B-fncl-2.IQ3_XS.gguf	IQ3_XS	0.32GB
Qwen2-0.5B-fncl-2.IQ3_S.gguf	IQ3_S	0.32GB
Qwen2-0.5B-fncl-2.Q3_K_S.gguf	Q3_K_S	0.32GB
Qwen2-0.5B-fncl-2.IQ3_M.gguf	IQ3_M	0.32GB
Qwen2-0.5B-fncl-2.Q3_K.gguf	Q3_K	0.33GB
Qwen2-0.5B-fncl-2.Q3_K_M.gguf	Q3_K_M	0.33GB
Qwen2-0.5B-fncl-2.Q3_K_L.gguf	Q3_K_L	0.34GB
Qwen2-0.5B-fncl-2.IQ4_XS.gguf	IQ4_XS	0.33GB
Qwen2-0.5B-fncl-2.Q4_0.gguf	Q4_0	0.33GB
Qwen2-0.5B-fncl-2.IQ4_NL.gguf	IQ4_NL	0.33GB
Qwen2-0.5B-fncl-2.Q4_K_S.gguf	Q4_K_S	0.36GB
Qwen2-0.5B-fncl-2.Q4_K.gguf	Q4_K	0.37GB
Qwen2-0.5B-fncl-2.Q4_K_M.gguf	Q4_K_M	0.37GB
Qwen2-0.5B-fncl-2.Q4_1.gguf	Q4_1	0.35GB
Qwen2-0.5B-fncl-2.Q5_0.gguf	Q5_0	0.37GB
Qwen2-0.5B-fncl-2.Q5_K_S.gguf	Q5_K_S	0.38GB
Qwen2-0.5B-fncl-2.Q5_K.gguf	Q5_K	0.39GB
Qwen2-0.5B-fncl-2.Q5_K_M.gguf	Q5_K_M	0.39GB
Qwen2-0.5B-fncl-2.Q5_1.gguf	Q5_1	0.39GB
Qwen2-0.5B-fncl-2.Q6_K.gguf	Q6_K	0.47GB
Qwen2-0.5B-fncl-2.Q8_0.gguf	Q8_0	0.49GB

Original model description:

license: apache-2.0 datasets: - haripritam/function-calling-alpaca

Further finetuned for 1 epoch taking the Qwen2-0.5B-fncl better function calling capabilities I think

but doesnot reject any function call i.e hallucinates functions even if they are not there.

Made a only-rejections dataset and a combined dataset finetuning on one epoch on both did not work. Dont know why.

Still this is the best extra small function calling model out there.

prompt template

prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
You are a helpful assistant with access to the following functions. Use them if required - { "name": "calculate_median", "description": "Calculate the median of a list of numbers", "parameters": { "type": "object", "properties": { "numbers": { "type": "array", "items": { "type": "number" }, "description": "The list of numbers" } }, "required": [ "numbers" ] } }

### Input:
Hi, I have a list of numbers and I need to find the median. The numbers are 5, 2, 9, 1, 7, 4, 6, 3, 8.

### Response:
"""