🚀 Blurdus-7b-v0.1
Blurdus-7b-v0.1 是一個通過合併多個模型而得到的模型。它藉助 LazyMergekit 工具,將多個基礎模型進行合併,在文本生成任務上展現出了一定的性能。
🚀 快速開始
安裝依賴
!pip install -qU transformers accelerate
運行示例代碼
from transformers import AutoTokenizer
import transformers
import torch
model = "222gate/Blurdus-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"])
✨ 主要特性
Blurdus-7b-v0.1 模型通過合併以下模型而成:
📦 安裝指南
運行以下命令安裝所需依賴:
!pip install -qU transformers accelerate
💻 使用示例
基礎用法
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "222gate/Blurdus-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"])
📚 詳細文檔
🧩 配置信息
models:
- model: 222gate/Blurred-Beagle-7b-slerp
parameters:
density: [1, 0.7, 0.1]
weight: 1.0
- model: 222gate/BrurryDog-7b-v0.1
parameters:
density: 0.5
weight: [0, 0.3, 0.7, 1]
- model: liminerity/Blur-7b-v1.21
parameters:
density: 0.33
weight:
- filter: mlp
value: 0.5
- value: 0
merge_method: ties
base_model: udkai/Turdus
parameters:
normalize: true
int8_mask: true
dtype: float16
詳細結果可查看 此處
指標 |
值 |
平均值 |
74.35 |
AI2 推理挑戰 (25 次少樣本學習) |
72.27 |
HellaSwag (10 次少樣本學習) |
88.50 |
MMLU (5 次少樣本學習) |
64.82 |
TruthfulQA (0 次少樣本學習) |
69.72 |
Winogrande (5 次少樣本學習) |
82.95 |
GSM8k (5 次少樣本學習) |
67.85 |
📄 許可證
本項目採用 Apache-2.0 許可證。