47 lines
1.3 KiB
Markdown
47 lines
1.3 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- mlabonne/mini-platypus
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
---
|
||
|
|
# 🦙🧠 Miniplatypus-7b
|
||
|
|
|
||
|
|
<center><img src="https://i.imgur.com/VkGvQym.png" width="300"></center>
|
||
|
|
|
||
|
|
This is a `Llama-2-7b-chat` model fine-tuned using QLoRA (4-bit precision) on the [`mlabonne/guanaco-llama2-1k`](https://huggingface.co/datasets/mlabonne/mini-platypus) dataset, which is a subset of the [`garage-bAInd/Open-Platypus`](https://huggingface.co/datasets/garage-bAInd/Open-Platypus).
|
||
|
|
|
||
|
|
## 🔧 Training
|
||
|
|
|
||
|
|
It was trained on a Google Colab notebook with a T4 GPU. It is mainly designed for educational purposes, not for inference.
|
||
|
|
|
||
|
|
## 💻 Usage
|
||
|
|
|
||
|
|
``` python
|
||
|
|
# pip install transformers accelerate
|
||
|
|
|
||
|
|
from transformers import AutoTokenizer
|
||
|
|
import transformers
|
||
|
|
import torch
|
||
|
|
|
||
|
|
model = "mlabonne/llama-2-7b-miniplatypus"
|
||
|
|
prompt = "What is a large language model?"
|
||
|
|
|
||
|
|
tokenizer = AutoTokenizer.from_pretrained(model)
|
||
|
|
pipeline = transformers.pipeline(
|
||
|
|
"text-generation",
|
||
|
|
model=model,
|
||
|
|
torch_dtype=torch.float16,
|
||
|
|
device_map="auto",
|
||
|
|
)
|
||
|
|
|
||
|
|
sequences = pipeline(
|
||
|
|
f'<s>[INST] {prompt} [/INST]',
|
||
|
|
do_sample=True,
|
||
|
|
top_k=10,
|
||
|
|
num_return_sequences=1,
|
||
|
|
eos_token_id=tokenizer.eos_token_id,
|
||
|
|
max_length=200,
|
||
|
|
)
|
||
|
|
for seq in sequences:
|
||
|
|
print(f"Result: {seq['generated_text']}")
|
||
|
|
```
|