OpenDolphinHermes_Llama2_7B

mergekit SLERP of these two models

• cognitivecomputations/dolphin-llama2-7b
• Tensoic/Llama-2-openhermes

🧩 Configuration

slices:
  - sources:
      - model: cognitivecomputations/dolphin-llama2-7b
        layer_range: [0, 32]
      - model: Tensoic/Llama-2-openhermes
        layer_range: [0, 32]
merge_method: slerp
base_model: Tensoic/Llama-2-openhermes
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

Prompt Template (ChatML)

<|im_start|>system
You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.
Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content.
Please ensure that your responses are socially unbiased and positive in nature.

If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct.
If you don't know the answer to a question, please don't share false information.
<|im_end|>
<|im_start|>user
{ .Prompt}
<|im_end|>
<|im_start|>assistant

OpenLLM Leaderboard

T	Model	Average	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K
0	meta-llama/llama-2-13b-hf	55.69	59.39	82.13	55.77	37.38	76.64	22.82
1	sethuiyer/OpenDolphinHermes_Llama2_7B	54.24	55.03	78.74	52.25	46.1	73.16	20.17
2	togethercomputer/Llama-2-7B-32K-Instruct	50.02	51.11	78.51	46.11	44.86	73.88	5.69
3	togethercomputer/LLaMa-2-7B-32K	47.07	47.53	76.14	43.33	39.23	71.9	4.32

Why?

I wanted a LLaMa2-7B model which is as good as base LLaMa2-13B model.

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "sethuiyer/OpenDolphinHermes_Llama2_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"])

Output:

A large language model is a type of artificial intelligence system that has been trained on a massive amount of data, often millions or even billions of words, to learn the patterns and relationships between words and phrases.
These models can then be used to generate new text, understand and translate languages, and perform various natural language processing tasks.
They have become increasingly popular in recent years due to advances in machine learning technology and their ability to achieve high levels of accuracy and performance on natural language processing tasks.
Examples of large language models include GPT-2, BERT, and T5.

Thanks

Thanks to Google Colab for the compute.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	54.24
AI2 Reasoning Challenge (25-Shot)	55.03
HellaSwag (10-Shot)	78.74
MMLU (5-Shot)	52.25
TruthfulQA (0-shot)	46.10
Winogrande (5-shot)	73.16
GSM8k (5-shot)	20.17