mvpmaster_-_kellemar-KrishnaHercules-0.1-7b-slerp-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

kellemar-KrishnaHercules-0.1-7b-slerp - bnb 8bits

• Model creator: https://huggingface.co/mvpmaster/
• Original model: https://huggingface.co/mvpmaster/kellemar-KrishnaHercules-0.1-7b-slerp/

Original model description:

license: apache-2.0 tags:

• merge
• mergekit
• lazymergekit
• Kukedlc/NeuralKrishna-7B-V2-DPO
• Locutusque/ChatHercules-2.5-Mistral-7B-DPO base_model:
• Kukedlc/NeuralKrishna-7B-V2-DPO
• Locutusque/ChatHercules-2.5-Mistral-7B-DPO

---

kellemar-KrishnaHercules-0.1-slerp

kellemar-KrishnaHercules-0.1-slerp is a merge of the following models using LazyMergekit:

• Kukedlc/NeuralKrishna-7B-V2-DPO
• Locutusque/ChatHercules-2.5-Mistral-7B-DPO

🧩 Configuration

models:
  - model: decruz07/kellemar-DPO-Orca-Distilled-7B-SLERP
    # No parameters necessary for base model
  - model: Kukedlc/NeuralKrishna-7B-V2-DPO
    parameters:
      density: 0.53
      weight: 0.4
  - model: Locutusque/ChatHercules-2.5-Mistral-7B-DPO
    parameters:
      density: 0.53
      weight: 0.4
merge_method: dare_ties
base_model: decruz07/kellemar-DPO-Orca-Distilled-7B-SLERP
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mvpmaster/kellemar-KrishnaHercules-0.1-slerp"
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"])