language:
- en
license: apache-2.0
Model Details
Zion_Alpha is the first REAL Hebrew model in the world. It wasn't finetuned for any tasks yet, but it actually understands and comprehends Hebrew. It can even do some decent translation. I've tested GPT4 vs Zion_Alpha, and out of the box, Zion_Alpha did a better job translating. I did the finetune using SOTA techniques and using my insights from years of underwater basket weaving. If you wanna offer me a job, just add me on Facebook.
Model architecture
Based on Mistral 7B. I didn't even bother to alter the tokenizer.
The recommended prompt setting is Debug-deterministic:
temperature: 1
top_p: 1
top_k: 1
typical_p: 1
min_p: 1
repetition_penalty: 1
The recommended instruction template is Mistral:
{%- for message in messages %}
{%- if message['role'] == 'system' -%}
{{- message['content'] -}}
{%- else -%}
{%- if message['role'] == 'user' -%}
{{-'[INST] ' + message['content'].rstrip() + ' [/INST]'-}}
{%- else -%}
{{-'' + message['content'] + '</s>' -}}
{%- endif -%}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{-''-}}
{%- endif -%}
English to hebrew example:
English to hebrew example:
History
The model was originally trained about 2 month after Mistral (v0.1) was released.
As of 04 June 2024, Zion_Alpha got the Highest SNLI score in the world among open source models in Hebrew, surpassing most of the models by a huge margin. (84.05 score)
Support
- My Ko-fi page ALL donations will go for research resources and compute, every bit counts 🙏🏻
- My Patreon ALL donations will go for research resources and compute, every bit counts 🙏🏻