File size: 1,198 Bytes
3647464
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
058ab3a
3647464
058ab3a
3647464
 
 
 
8f0cf73
3647464
 
 
 
 
8a91ea4
 
3647464
 
 
 
 
058ab3a
3647464
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
---
language: en
tags:
- en
- english
- gpt2
- gpt3
- text-generation
- lm
- nlp
datasets:
- cnn_dailymail
widget:
- text: "Ever noticed how plane seats appear to be getting smaller and smaller? "

inference:
  parameters:
    max_length: 120
    do_sample: True
    temperature: 0.8
---

# GPT-3 small

Pretrained GPT-3 small, continuing the development of GPT NEO, with architecture that purposefully mimics that of GPT-3, model was trained on CNN Daily Mail News dataset for text generation.
# How to use the model

~~~~
from transformers import GPT2Tokenizer, GPTNeoForCausalLM

tokenizer = GPT2Tokenizer.from_pretrained('gpt3-small-finetune-cnndaily-news')
model = GPTNeoForCausalLM.from_pretrained('gpt3-small-finetune-cnndaily-news')

text = "Ever noticed how plane seats appear to be getting smaller and smaller? "
input_ids = tokenizer.encode(text, return_tensors='pt')
max_length = 150

sample_outputs = model.generate(input_ids, do_sample=True, max_length=max_length,temperature = 0.8)

for i, sample_output in enumerate(sample_outputs):
    print(">> Generated text {}\n\n{}".format(i+1, tokenizer.decode(sample_output.tolist())))
    print('\n---')
~~~~


## Author
`
Phan Minh Toan 
`