Text2Text Generation
Transformers
PyTorch
t5
text-generation-inference
Inference Endpoints
chiayewken commited on
Commit
69557a0
1 Parent(s): 0a4814f

Create model card

Browse files
Files changed (1) hide show
  1. README.md +45 -0
README.md ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - tatsu-lab/alpaca
5
+ ---
6
+
7
+ ## 🍮 🦙 Flan-Alpaca: Instruction Tuning from Humans and Machines
8
+
9
+ Our [repository](https://github.com/declare-lab/flan-alpaca) contains code for extending the [Stanford Alpaca](https://github.com/tatsu-lab/stanford_alpaca)
10
+ synthetic instruction tuning to existing instruction-tuned models such as [Flan-T5](https://arxiv.org/abs/2210.11416).
11
+ The pretrained models and demos are available on HuggingFace 🤗 :
12
+ [Base](https://huggingface.co/declare-lab/flan-alpaca-base) (220M),
13
+ [Large](https://huggingface.co/declare-lab/flan-alpaca-large) (770M),
14
+ [XL](https://huggingface.co/declare-lab/flan-alpaca-xl) (3B),
15
+ XXL (11B, Coming soon)
16
+
17
+ ### Why?
18
+
19
+ [Alpaca](https://crfm.stanford.edu/2023/03/13/alpaca.html) represents an exciting new direction
20
+ to approximate the performance of large language models (LLMs) like ChatGPT cheaply and easily.
21
+ Concretely, they leverage an LLM such as GPT-3 to generate instructions as synthetic training data.
22
+ The synthetic data which covers more than 50k tasks can then be used to finetune a smaller model.
23
+ However, the original implementation is less accessible due to licensing constraints of the
24
+ underlying [LLaMA](https://ai.facebook.com/blog/large-language-model-llama-meta-ai/) model.
25
+ Furthermore, users have noted [potential noise](https://github.com/tloen/alpaca-lora/issues/65) in the synthetic
26
+ dataset. Hence, it may be better to explore a fully accessible model that is already trained on high-quality (but
27
+ less diverse) instructions such as [Flan-T5](https://arxiv.org/abs/2210.11416).
28
+
29
+ ### Usage
30
+
31
+ ```
32
+ from transformers import pipeline
33
+
34
+ prompt = "Write an email about an alpaca that likes flan"
35
+ model = pipeline(model="declare-lab/flan-alpaca-xl")
36
+ model(prompt, max_length=128, do_sample=True)
37
+
38
+ # Dear AlpacaFriend,
39
+ # My name is Alpaca and I'm 10 years old.
40
+ # I'm excited to announce that I'm a big fan of flan!
41
+ # We like to eat it as a snack and I believe that it can help with our overall growth.
42
+ # I'd love to hear your feedback on this idea.
43
+ # Have a great day!
44
+ # Best, AL Paca
45
+ ```