Downloads · 30 days
6
11% of all-time downloads
jeff-bezos-zh-gemma-4-e4b-merged
xihajun/jeff-bezos-zh-gemma-4-e4b-merged
jeff-bezos-zh-gemma-4-e4b-merged is a text generation model from xihajun. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as gemma.
Chinese Jeff Bezos style persona model based on Gemma 4 E4B, merged from a LoRA adapter trained at maxlength 1536.
All-time downloads
56
Public
Parameters
7.9B
15.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors15.9 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- gemma
- Model type
- gemma4
- Access
- Public
- Created
- May 12, 2026
- Updated
- May 12, 2026
- SHA
- 44de10de
Base models
Model card
From the Hugging Face model README
jeff-bezos-zh-gemma-4-e4b-merged
Chinese Jeff Bezos style persona model based on Gemma 4 E4B, merged from a LoRA adapter trained at max_length 1536.
This is a merged Transformers model based on google/gemma-4-E4B-it plus a PEFT LoRA adapter. It is intended for experimental persona and style transfer evaluation.
Intended Use
- Text-generation experiments with Chinese and English prompts.
- Server-side validation with Hugging Face Transformers.
- Conversion experiments for mobile runtimes such as LiteRT-LM.
Source
- Base model:
google/gemma-4-E4B-it - Adapter source path on training host:
/home/ubuntu/gemma4_e4b_jeff_bezos_zh_lora_1536_full2_may09 - Merge format: full Transformers safetensors model with tokenizer/processor files.
Notes
This repository is not published by Google. Validate model quality, safety, and deployment behavior for your own use case.
Gemma is a trademark of Google LLC.
- Task
- Text Generation
- Library
- transformers
- Type
- gemma4
- License
- gemma
- Languages
- zh, en
- Created
- May 12, 2026
- Updated
- May 12, 2026