metadata

inference: false
tags:
  - merge
  - mergekit
  - lazymergekit
base_model:
  - senseable/Westlake-7B
  - Guilherme34/Samantha-v2
  - uukuguy/speechless-mistral-six-in-one-7b
license: apache-2.0
language:
  - en
library_name: transformers
pipeline_tag: conversational

Nandine-7b

Nandine

Nandine-7b is a merge of the following models using LazyMergekit:

Nandine-7b represents a harmonious amalgamation of narrative skill, empathetic interaction, intellectual depth, and eloquent communication.

Key Features:

Narrative Excellence: Drawing 55% of its prowess from Westlake-7B, Nandine-7b excels in crafting compelling narratives and engaging character dialogues.
Diverse Expertise: Benefits from the 35% contribution of speechless-mistral-six-in-one-7b, integrating insights from six advanced Mistral-7B models, covering areas like cognitive science, emotional intelligence, and social dynamics.
Human Connection: Gains a 10% influence from Samantha-v2, fostering a warm, approachable demeanor, enhancing user interaction quality while adhering to ethical guidelines.

Intended Use: Ideal for users seeking a versatile AI companion for creative writing, thoughtful discussions, and general assistance. Nandine-7b aims to provide a delightful, insightful, and emotionally resonant user experience.

Limitations: While Nandine-7b excels in various domains, it may not perfectly replicate human nuances in certain complex emotional or contextual scenarios. Users should be aware of these limitations in highly specialized or sensitive topics.

🧩 Configuration

models:
  - model: senseable/Westlake-7B
    parameters:
      weight: 0.55
      density: 0.6
  - model: Guilherme34/Samantha-v2
    parameters:
      weight: 0.10
      density: 0.3
  - model: uukuguy/speechless-mistral-six-in-one-7b
    parameters:
      weight: 0.35
      density: 0.6
merge_method: dare_ties
base_model: mistralai/Mistral-7B-v0.1
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "sethuiyer/Nandine-7b"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])