Based on Gemma-2-2B-it, enhanced with multiple tuning techniques to improve its general performance.
2B
148 Pulls Updated 3 weeks ago
Updated 3 weeks ago
3 weeks ago
ccfbd3d3f538 · 2.8GB
model
archgemma2
·
parameters2.61B
·
quantizationQ8_0
2.8GB
template
<bos><start_of_turn>user
{{ if .System }}{{ .System }} {{ end }}{{ .Prompt }}<end_of_turn>
<start_of_turn>model
{{ .Response }}<end_of_turn><eos>
146B
params
{"repeat_penalty":1,"stop":["<start_of_turn>","<end_of_turn>","<bos>","<eos>"]}
120B
license
This model is based on Gemma-2-2B-it and is subject to the Gemma Terms of Use. For detailed information, please refer to the official Gemma license page.
153B
Readme
This model is based on Gemma-2-2B-it, enhanced with multiple tuning techniques to improve its general performance. While it excels in Japanese language tasks, it’s designed to meet diverse needs globally.
Links
Original:
GGUF:
Reference:
Modelfile
FROM EZO-Common-T2-2B-gemma-2-it.Q8_0.gguf
TEMPLATE """<bos><start_of_turn>user
{{ if .System }}{{ .System }} {{ end }}{{ .Prompt }}<end_of_turn>
<start_of_turn>model
{{ .Response }}<end_of_turn><eos>
"""
PARAMETER repeat_penalty 1
PARAMETER stop <start_of_turn>
PARAMETER stop <end_of_turn>
PARAMETER stop <bos>
PARAMETER stop <eos>
LICENSE "This model is based on Gemma-2-2B-it and is subject to the Gemma Terms of Use. For detailed information, please refer to the official Gemma license page."