sharpbai commited on
Commit
6c2eedc
1 Parent(s): 69a3b6c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -0
README.md CHANGED
@@ -8,3 +8,46 @@ inference: false
8
 
9
  A 405M split weight version of [lmsys/vicuna-7b-v1.3](https://huggingface.co/lmsys/vicuna-7b-v1.3)
10
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
 
9
  A 405M split weight version of [lmsys/vicuna-7b-v1.3](https://huggingface.co/lmsys/vicuna-7b-v1.3)
10
 
11
+ Original model is down below
12
+
13
+ # Vicuna Model Card
14
+
15
+ ## Model Details
16
+
17
+ Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT.
18
+
19
+ - **Developed by:** [LMSYS](https://lmsys.org/)
20
+ - **Model type:** An auto-regressive language model based on the transformer architecture.
21
+ - **License:** Non-commercial license
22
+ - **Finetuned from model:** [LLaMA](https://arxiv.org/abs/2302.13971).
23
+
24
+ ### Model Sources
25
+
26
+ - **Repository:** https://github.com/lm-sys/FastChat
27
+ - **Blog:** https://lmsys.org/blog/2023-03-30-vicuna/
28
+ - **Paper:** https://arxiv.org/abs/2306.05685
29
+ - **Demo:** https://chat.lmsys.org/
30
+
31
+ ## Uses
32
+
33
+ The primary use of Vicuna is research on large language models and chatbots.
34
+ The primary intended users of the model are researchers and hobbyists in natural language processing, machine learning, and artificial intelligence.
35
+
36
+ ## How to Get Started with the Model
37
+
38
+ Command line interface: https://github.com/lm-sys/FastChat#vicuna-weights.
39
+ APIs (OpenAI API, Huggingface API): https://github.com/lm-sys/FastChat/tree/main#api.
40
+
41
+ ## Training Details
42
+
43
+ Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning.
44
+ The training data is around 140K conversations collected from ShareGPT.com.
45
+ See more details in the "Training Details of Vicuna Models" section in the appendix of this [paper](https://arxiv.org/pdf/2306.05685.pdf).
46
+
47
+ ## Evaluation
48
+
49
+ Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this [paper](https://arxiv.org/pdf/2306.05685.pdf).
50
+
51
+ ## Difference between different versions of Vicuna
52
+ See [vicuna_weights_version.md](https://github.com/lm-sys/FastChat/blob/main/docs/vicuna_weights_version.md)
53
+