NeuralSynthesis-7B-v0.1
NeuralSynthesis-7B-v0.1 is a merge of the following models using LazyMergekit:
🧩 Configuration
models:
- model: MiniMoog/Mergerix-7b-v0.3
- model: automerger/Ognoexperiment27Multi_verse_model-7B
- model: AurelPx/Percival_01-7b-slerp
- model: automerger/YamshadowExperiment28-7B
merge_method: model_stock
base_model: automerger/YamshadowExperiment28-7B
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Kukedlc/NeuralSynthesis-7B-v0.1"
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"])
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 76.80 |
| AI2 Reasoning Challenge (25-Shot) | 73.04 |
| HellaSwag (10-Shot) | 89.18 |
| MMLU (5-Shot) | 64.37 |
| TruthfulQA (0-shot) | 78.15 |
| Winogrande (5-shot) | 85.24 |
| GSM8k (5-shot) | 70.81 |
- Downloads last month
- 9,779
Model tree for Kukedlc/NeuralSynthesis-7B-v0.1
Spaces using Kukedlc/NeuralSynthesis-7B-v0.1 10
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard73.040
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard89.180
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard64.370
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard78.150
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard85.240
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard70.810