RichardErkhov/aloobun_-_d-Qwen2-0.5B-gguf

Quantization made by Richard Erkhov.

d-Qwen2-0.5B - GGUF

Model creator: https://huggingface.co/aloobun/
Original model: https://huggingface.co/aloobun/d-Qwen2-0.5B/

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

Original model description:

license: apache-2.0 library_name: transformers tags: - qwen2 - distillation datasets: - EleutherAI/the_pile_deduplicated

This is a distillation experiment with Qwen2-1.5B as teacher and Qwen2-0.5B as student model respectively.
Samples were taken from the Pile dataset.
optimizer: SM3, scheduler: cosine with warmup, lr=2e-5

Qwen2 is the new series of Qwen large language models. For Qwen2, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters, including a Mixture-of-Experts model. This repo contains distilled 0.5B Qwen2 language model.