<img src=https://huggingface.co/lodrick-the-lafted/Grafted-Hermetic-Platypus-D-2x7B/resolve/main/ghp.png>
Grafted-Hermetic-Platypus-D-2x7B
MoE merge of
Basically the same thing as GHP-A.
Prompt Format
Both the default Mistral-Instruct tags and Alpaca are fine, so either:
<s>[INST] {sys_prompt} {instruction} [/INST]
or
{sys_prompt}
### Instruction:
{instruction}
### Response:
The tokenizer default is Alpaca this time around.
Usage
from transformers import AutoTokenizer
import transformers
import torch
model = "lodrick-the-lafted/Grafted-Hermetic-Platypus-A-2x7B"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.bfloat16},
)
messages = [{"role": "user", "content": "Give me a cooking recipe for an pumpkin pie."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_p=0.95)
print(outputs[0]["generated_text"])
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 64.24 |
| AI2 Reasoning Challenge (25-Shot) | 58.87 |
| HellaSwag (10-Shot) | 82.89 |
| MMLU (5-Shot) | 61.96 |
| TruthfulQA (0-shot) | 61.02 |
| Winogrande (5-shot) | 77.43 |
| GSM8k (5-shot) | 43.29 |