Downloads · 30 days
54
1% of all-time downloads
seyf1elislam/WestKunai-Hermes-10.7b-test
WestKunai-Hermes-10.7b-test is a text generation model from seyf1elislam. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as cc-by-nc-4.0.
Replicate the configuration utilized in the froggeric/WestLake-10.7B-v2 model to extend the WestKunai-Hermes-7b model to 10.7b.
Downloads · 30 days
54
1% of all-time downloads
All-time downloads
8.6K
Public
Parameters
10.7B
21.5 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors21.5 GB · 100%
From the Hugging Face model README
Replicate the configuration utilized in the froggeric/WestLake-10.7B-v2 model to extend the WestKunai-Hermes-7b model to 10.7b.
| Metric | diff | Current(10.7b) | Origin(7b) |
|---|---|---|---|
| Avg. | <span style="color:red">-3.76</span> | 69.75 | 73.51 |
| AI2 Reasoning Challenge (25-Shot) | <span style="color:red">-3.07</span> | 68.09 | 71.16 |
| HellaSwag (10-Shot) | <span style="color:red">-0.66</span> | 87.10 | 87.76 |
| MMLU (5-Shot) | <span style="color:red">-0.34</span> | 64.43 | 64.77 |
| TruthfulQA (0-shot) | <span style="color:red">-0.97</span> | 64.28 | 65.25 |
| Winogrande (5-shot) | <span style="color:red">-0.31</span> | 82.72 | 83.03 |
| GSM8k (5-shot) | <span style="color:red">-17.21</span> | 51.86 | 69.07 |
The following models were included in the merge:
The following YAML configuration was used to produce this model:
slices:
- sources:
- model: seyf1elislam/WestKunai-Hermes-7b
layer_range: [0,9]
- sources:
- model: seyf1elislam/WestKunai-Hermes-7b
layer_range: [5,14]
- sources:
- model: seyf1elislam/WestKunai-Hermes-7b
layer_range: [10,19]
- sources:
- model: seyf1elislam/WestKunai-Hermes-7b
layer_range: [15,24]
- sources:
- model: seyf1elislam/WestKunai-Hermes-7b
layer_range: [20,32]
merge_method: passthrough
dtype: bfloat16
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "seyf1elislam/WestKunai-Hermes-10.7b-test"
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"])
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 69.75 |
| AI2 Reasoning Challenge (25-Shot) | 68.09 |
| HellaSwag (10-Shot) | 87.10 |
| MMLU (5-Shot) | 64.43 |
| TruthfulQA (0-shot) | 64.28 |
| Winogrande (5-shot) | 82.72 |
| GSM8k (5-shot) | 51.86 |