File size: 1,264 Bytes
a5c4750
 
 
 
 
 
 
 
 
 
 
34c3f9e
a5c4750
 
 
 
 
 
 
 
 
 
 
6420f7d
 
 
 
 
01e969d
6420f7d
 
 
 
 
 
 
 
 
 
34c3f9e
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
---
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
base_model: unsloth/llama-3-8b-bnb-4bit
pipeline_tag: text-generation
---

# Uploaded  model

- **Developed by:** underscore2
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-3-8b-bnb-4bit

This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)

# Usage
```
from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
        model_name = "underscore2/llama3-8b-mlsubs",
        max_seq_length = max_seq_length,
        dtype = dtype,
        load_in_4bit = load_in_4bit,
    )
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer("[POST START]: New Architecture that replaces the MLP by using literal magic", return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 1000, repetition_penalty=1.4)
```