Just a place to store some of the Jupyter and Colab notebooks I've used when playing with LLMs.
Here's a quick spreadsheet I've been using to summarize the difference of the models. Feel free to add comments or suggestions. https://docs.google.com/spreadsheets/d/1RftE-WMHCQF82DFJPwh-ch8FIE4jEFClIfy8C9enMho/edit#gid=0
Models I have tried:
- stabilityai/japanese-stablelm-base-alpha-7b
- stabilityai/japanese-stablelm-instruct-alpha-7b
- weblab-10b
- qwen-7b
- rinna/bilingual-gpt-neox-4b
- rinna/bilingual-gpt-neox-4b-instruction-sft
- rinna/bilingual-gpt-neox-4b-instruction-ppo
- line-corporation/japanese-large-lm-3.6b
- line-corporation/japanese-large-lm-3.6b-instruction-sft
- ELYZA-japanese-Llama-2-7b-fast-instruct