L3.1-70b-Ginny

L3.1-70b-Ginny

Like with everything, I have to start somewhere right? As such this model is named Ginny.

L3.1-70b-Ginny is a merge of the following models using LazyMergekit:

• NousResearch/Hermes-3-Llama-3.1-70B
• Fizzarolli/L3.1-70b-glitz-v0.2
• cyberagent/Llama-3.1-70B-Japanese-Instruct-2407
• Sao10K/L3-70B-Euryale-v2.1

Using Hermes as a base, I mixed in Glitz and Euryale which I both liked. I think I prefer Glitz more actually.

Additionally I decided to throw in cyberagent's Japanese Instruct in the hopes it will boost Japanese capabilities.

(Though on recommendations from others, I've steeled myself to never use Hermes as base ever again.)

Model Testing

• The Japanese Instruct was a success in my books. The model appears to be more capable in Japanese and I think it can do translations given the right prompt.
• Model can get quite rambly. Lower Temp helps.

Quants & Hosts

Yap / Chat Format

Hermes likes ChatML, but the other 3 models use L3 Instruct.

...Use L3 Instruct.

🧩 Configuration

models:
  - model: NousResearch/Hermes-3-Llama-3.1-70B
    parameters:
      density: 0.33
      weight: 0.25
  - model: Fizzarolli/L3.1-70b-glitz-v0.2
    parameters:
      density: 0.7
      weight: 0.5
  - model: cyberagent/Llama-3.1-70B-Japanese-Instruct-2407
    parameters:
      density: 0.5
      weight: 0.25
  - model: Sao10K/L3-70B-Euryale-v2.1
    parameters:
      density: 0.7
      weight: 0.5

merge_method: ties
base_model: NousResearch/Hermes-3-Llama-3.1-70B
parameters:
  normalize: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "KaraKaraWitch/L3.1-70b-Ginny"
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"])