File size: 1,293 Bytes
f4272d4
 
 
 
 
 
 
 
 
 
 
c952f92
f4272d4
 
5c05a27
f4272d4
4fb59f9
 
 
f4272d4
4fb59f9
f4272d4
5c05a27
4fb59f9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model: unsloth/mistral-7b-v0.3-bnb-4bit
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- mistral
- trl
- sft
---

# Jokestral

This model was created by fine-tuning `unsloth/mistral-7b-v0.3-bnb-4bit` on [Short jokes dataset](https://www.kaggle.com/datasets/abhinavmoudgil95/short-jokes).
So the only purpose of this model is the generation of cringe jokes. </br>
Just write the first few words and get your joke.

# Usage

[**Goodle Colab example**](https://colab.research.google.com/drive/13N1O-fq-vjr8FUrsUU6f24fPpyf0ZwOS#scrollTo=UBSG1UTV85Vq)

```
pip install transformers
pip install --no-deps "trl<0.9.0" peft accelerate bitsandbytes
```
```
from transformers import AutoTokenizer,AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained("SantaBot/Jokestral_4bit",)
tokenizer = AutoTokenizer.from_pretrained("SantaBot/Jokestral_4bit")

inputs = tokenizer(
[
    "My doctor" # YOUR PROMPT HERE
], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)
```

**The output should be something like** : </br>
`['<s> My doctor told me I have to stop m4sturb4t1ng. I asked him why and he said ""Because I\'m trying to examine you.""\n</s>']`