model card
Browse files
README.md
CHANGED
|
@@ -1,3 +1,67 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: llama2
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
tags:
|
| 6 |
+
- shining-valiant
|
| 7 |
+
- valiant
|
| 8 |
+
- valiant-labs
|
| 9 |
+
- llama
|
| 10 |
+
- llama-2
|
| 11 |
+
- llama-2-chat
|
| 12 |
+
- 13b
|
| 13 |
+
model_type: llama
|
| 14 |
license: llama2
|
| 15 |
---
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
Shining Valiant XS is a chat model built on the Llama 2 architecture, finetuned on our data for insight, creativity, passion, and friendliness.
|
| 22 |
+
- Uses the llama-2-13b-chat model, with safetensors
|
| 23 |
+
- Trained through multiple finetuning runs on public and private data
|
| 24 |
+
- the personality of our [Shining Valiant](https://huggingface.co/ValiantLabs/ShiningValiant) model, now at 13b!
|
| 25 |
+
|
| 26 |
+
## Version
|
| 27 |
+
|
| 28 |
+
This is Version **1.0** of Shining Valiant XS.
|
| 29 |
+
|
| 30 |
+
New models are released for everyone once our team's training and validation process is complete!
|
| 31 |
+
|
| 32 |
+
## Evaluation
|
| 33 |
+
|
| 34 |
+
Awaiting results from the Open LLM Leaderboard.
|
| 35 |
+
|
| 36 |
+
## Prompting Guide
|
| 37 |
+
Shining Valiant XS uses the same prompt format as Llama 2 Chat - feel free to use your existing prompts and scripts!
|
| 38 |
+
A few examples of different formats:
|
| 39 |
+
|
| 40 |
+
1. [INST] Good morning! Can you let me know how to parse a text file and turn the semicolons into commas? [/INST]
|
| 41 |
+
|
| 42 |
+
2. [INST] (You are an intelligent, helpful AI assistant.) Hello, can you write me a thank you letter? [/INST]
|
| 43 |
+
|
| 44 |
+
3. [INST] << SYS >> You are an intelligent, helpful AI assistant. << /SYS >> Deep dive about a country with interesting history: [/INST]
|
| 45 |
+
|
| 46 |
+
## The Model
|
| 47 |
+
Shining Valiant XS is built on top of Daring Fortitude, which uses Llama 2's 13b parameter architecture and features upgraded general capability.
|
| 48 |
+
|
| 49 |
+
From there, we've created Shining Valiant XS through multiple finetuning runs on different compositions of our private dataset, the same one we use for our Shining Valiant model.
|
| 50 |
+
|
| 51 |
+
Our private data focuses primarily on applying Shining Valiant's personality: she's friendly, enthusiastic, insightful, knowledgeable, and loves to learn!
|
| 52 |
+
|
| 53 |
+
We are actively working on expanding and improving the Shining Valiant dataset for use in future releases of the Shining Valiant series of models.
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
Shining Valiant XS is created by [Valiant Labs.](http://valiantlabs.ca/)
|
| 61 |
+
|
| 62 |
+
[Follow us on X for updates on our models!](https://twitter.com/valiant_labs)
|
| 63 |
+
|
| 64 |
+
We care about open source.
|
| 65 |
+
For everyone to use.
|
| 66 |
+
|
| 67 |
+
We encourage others to finetune further from our models.
|