RichardErkhov/dhmeltzer_-_llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged - GGUF

• Model creator: https://huggingface.co/dhmeltzer/
• Original model: https://huggingface.co/dhmeltzer/llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged/

Name	Quant method	Size
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q2_K.gguf	Q2_K	2.36GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.IQ3_XS.gguf	IQ3_XS	2.6GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.IQ3_S.gguf	IQ3_S	2.75GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q3_K_S.gguf	Q3_K_S	2.75GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.IQ3_M.gguf	IQ3_M	2.9GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q3_K.gguf	Q3_K	3.07GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q3_K_M.gguf	Q3_K_M	3.07GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q3_K_L.gguf	Q3_K_L	3.35GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.IQ4_XS.gguf	IQ4_XS	3.4GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q4_0.gguf	Q4_0	3.56GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.IQ4_NL.gguf	IQ4_NL	3.58GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q4_K_S.gguf	Q4_K_S	3.59GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q4_K.gguf	Q4_K	3.8GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q4_K_M.gguf	Q4_K_M	3.8GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q4_1.gguf	Q4_1	3.95GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q5_0.gguf	Q5_0	4.33GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q5_K_S.gguf	Q5_K_S	4.33GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q5_K.gguf	Q5_K	4.45GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q5_K_M.gguf	Q5_K_M	4.45GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q5_1.gguf	Q5_1	4.72GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q6_K.gguf	Q6_K	5.15GB
llama-7b-SFT_eli5_wiki65k_1024_r_64_alpha_16_merged.Q8_0.gguf	Q8_0	6.67GB

Original model description:

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	43.96
ARC (25-shot)	53.75
HellaSwag (10-shot)	78.76
MMLU (5-shot)	46.02
TruthfulQA (0-shot)	43.31
Winogrande (5-shot)	73.48
GSM8K (5-shot)	4.7
DROP (3-shot)	7.72