lemon-mint_-_gemma-7b-openhermes-v0.80-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

gemma-7b-openhermes-v0.80 - bnb 4bits

• Model creator: https://huggingface.co/lemon-mint/
• Original model: https://huggingface.co/lemon-mint/gemma-7b-openhermes-v0.80/

Original model description:

library_name: transformers language:

• en license: gemma tags:
• gemma
• pytorch
• instruct
• finetune base_model: google/gemma-1.1-7b-it pipeline_tag: text-generation datasets:
• teknium/OpenHermes-2.5

---

Gemma 7B OpenHermes v0.80

• Eval Loss: 0.4544
• Train Loss: 0.3129
• lr: 5e-5
• optimizer: adamw
• lr_scheduler_type: cosine

Model Details

This is an instruction-following model finetuned from the Gemma 1.1 7B model. It was finetuned on the OpenHermes-2.5 dataset to improve its ability to engage in open-ended conversation and respond helpfully to user instructions and queries. The model can engage in dialogue, answer questions, and assist with a variety of tasks.

Model Description

• Developed by: lemon-mint
• Model type: Gemma
• Language(s) (NLP): English
• License: gemma-terms-of-use
• Finetuned from model: google/gemma-1.1-7b-it

Limitations and Ethical Considerations

As Gemma 7B OpenHermes has been trained on extensive web data, biases present in the training data may be reflected in the model. Additionally, there is a possibility that it may generate sentences containing errors or incorrect information. Therefore, rather than blindly trusting the model's output, it is necessary to refer to it with caution.