L3-Urvashi-v2

L3-Urvashi-v2 is a merge of the following models using LazyMergekit:

• tannedbum/L3-Nymeria-v2-8B
• Divyansh008/L3-Urvashi-8B-Slerp
• ZeroXClem/Llama-3.1-8B-AthenaSky-MegaMix

🧩 Configuration

models:
  - model: tannedbum/L3-Nymeria-v2-8B
    parameters:
      weight: 1.0
  - model: Divyansh008/L3-Urvashi-8B-Slerp
    parameters:
      weight: 1.0
  - model: ZeroXClem/Llama-3.1-8B-AthenaSky-MegaMix
    parameters:
      weight: 1.0
merge_method: della
base_model: tannedbum/L3-Nymeria-v2-8B
parameters:
  density: 0.7
  lambda: 1.1
  epsilon: 0.2
tokenizer_source: union
chat_template: "auto"
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Divyansh008/L3-Urvashi-v2"
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"])