|
|
--- |
|
|
library_name: transformers |
|
|
tags: |
|
|
- fine-tuning |
|
|
- prose |
|
|
- KTO |
|
|
- axolotl |
|
|
- finetune |
|
|
- roleplaying |
|
|
- creative-writing |
|
|
base_model: |
|
|
- Delta-Vector/Rei-24B-Base |
|
|
--- |
|
|
<style> |
|
|
@import url('https://fonts.googleapis.com/css2?family=Share+Tech+Mono&display=swap'); |
|
|
|
|
|
body { |
|
|
font-family: 'Share Tech Mono', monospace; |
|
|
background: #121212; |
|
|
color: #E0E0E0; |
|
|
margin: 0; |
|
|
padding: 0; |
|
|
font-size: 16px; |
|
|
} |
|
|
.container { |
|
|
margin: 20px auto; |
|
|
background-color: #1E1E1E; |
|
|
padding: 20px; |
|
|
border-radius: 8px; |
|
|
box-shadow: 0 0 25px rgba(255, 0, 51, 0.3); |
|
|
border: 1px solid #FF0033; |
|
|
position: relative; |
|
|
max-width: 1000px; |
|
|
} |
|
|
.header h1 { |
|
|
font-size: 32px; |
|
|
color: #FF0033; |
|
|
margin: 0 0 20px 0; |
|
|
text-align: center; |
|
|
text-shadow: 0 0 10px rgba(255, 0, 51, 0.7); |
|
|
text-transform: uppercase; |
|
|
} |
|
|
.section { |
|
|
margin-top: 30px; |
|
|
} |
|
|
.section h2 { |
|
|
font-size: 26px; |
|
|
color: #FF0033; |
|
|
text-align: center; |
|
|
border-bottom: 1px solid #FF0033; |
|
|
padding-bottom: 10px; |
|
|
margin-bottom: 20px; |
|
|
} |
|
|
.info p { |
|
|
color: #E0E0E0; |
|
|
line-height: 1.7; |
|
|
font-size: 16px; |
|
|
} |
|
|
.info img { |
|
|
width: 100%; |
|
|
max-width: 600px; |
|
|
border-radius: 4px; |
|
|
margin: 0 auto 15px; |
|
|
display: block; |
|
|
box-shadow: 0 0 20px rgba(0, 0, 0, 0.5); |
|
|
border: 1px solid #444; |
|
|
} |
|
|
a { |
|
|
color: #FF0033; |
|
|
text-decoration: none; |
|
|
transition: color 0.2s ease, text-shadow 0.2s ease; |
|
|
} |
|
|
a:hover { |
|
|
color: #FFFFFF; |
|
|
text-shadow: 0 0 8px rgba(255, 0, 51, 0.8); |
|
|
} |
|
|
.button { |
|
|
display: inline-block; |
|
|
background-color: transparent; |
|
|
color: #FF0033; |
|
|
padding: 10px 20px; |
|
|
border-radius: 4px; |
|
|
cursor: pointer; |
|
|
text-decoration: none; |
|
|
border: 1px solid #FF0033; |
|
|
transition: background-color 0.2s ease, box-shadow 0.2s ease, color 0.2s ease; |
|
|
} |
|
|
.button:hover { |
|
|
background-color: #FF0033; |
|
|
color: #121212; |
|
|
box-shadow: 0 0 15px rgba(255, 0, 51, 0.7); |
|
|
} |
|
|
pre { |
|
|
background-color: #0d0d0d; |
|
|
padding: 15px; |
|
|
border-radius: 4px; |
|
|
overflow-x: auto; |
|
|
border: 1px solid #444; |
|
|
border-left: 3px solid #FF0033; |
|
|
color: #E0E0E0; |
|
|
} |
|
|
code { |
|
|
font-family: 'Share Tech Mono', monospace; |
|
|
color: #E0E0E0; |
|
|
} |
|
|
.info-card { |
|
|
background: #2a2a2a; |
|
|
border: 1px solid #444; |
|
|
border-radius: 8px; |
|
|
overflow: hidden; |
|
|
} |
|
|
.info-header { |
|
|
background: #1a1a1a; |
|
|
padding: 20px; |
|
|
border-bottom: 1px solid #444; |
|
|
} |
|
|
.info-header h3 { |
|
|
color: #FFFFFF; |
|
|
margin: 0 0 10px 0; |
|
|
font-size: 22px; |
|
|
text-align: center; |
|
|
} |
|
|
.model-tags { |
|
|
display: flex; |
|
|
justify-content: center; |
|
|
gap: 10px; |
|
|
flex-wrap: wrap; |
|
|
} |
|
|
.model-tag { |
|
|
background: #121212; |
|
|
color: #FF0033; |
|
|
padding: 5px 10px; |
|
|
border-radius: 4px; |
|
|
font-size: 12px; |
|
|
border: 1px solid #FF0033; |
|
|
text-transform: uppercase; |
|
|
} |
|
|
.model-composition { |
|
|
padding: 20px; |
|
|
border-bottom: 1px solid #444; |
|
|
} |
|
|
.model-composition h4 { |
|
|
color: #FFFFFF; |
|
|
margin: 0 0 15px 0; |
|
|
font-size: 18px; |
|
|
text-align: center; |
|
|
} |
|
|
.composition-list { |
|
|
list-style: none; |
|
|
padding: 0; |
|
|
margin: 0; |
|
|
display: grid; |
|
|
gap: 10px; |
|
|
} |
|
|
.composition-list li { |
|
|
color: #E0E0E0; |
|
|
display: flex; |
|
|
align-items: baseline; |
|
|
gap: 8px; |
|
|
} |
|
|
.model-component { |
|
|
font-weight: bold; |
|
|
min-width: 120px; |
|
|
} |
|
|
.model-description { |
|
|
padding: 20px; |
|
|
background: #2a2a2a; |
|
|
} |
|
|
.metrics-section { |
|
|
margin-bottom: 30px; |
|
|
} |
|
|
.metrics-section details { |
|
|
background: #2a2a2a; |
|
|
border: 1px solid #444; |
|
|
border-radius: 8px; |
|
|
padding: 15px; |
|
|
margin-bottom: 15px; |
|
|
} |
|
|
.metrics-section summary { |
|
|
color: #FF0033; |
|
|
font-size: 18px; |
|
|
cursor: pointer; |
|
|
outline: none; |
|
|
padding: 5px 0; |
|
|
text-align: center; |
|
|
font-weight: bold; |
|
|
} |
|
|
.creator-section { |
|
|
margin: 20px 0; |
|
|
} |
|
|
.creator-badge { |
|
|
display: inline-flex; |
|
|
align-items: center; |
|
|
background: #1a1a1a; |
|
|
border: 1px solid #444; |
|
|
border-radius: 4px; |
|
|
padding: 10px 15px; |
|
|
} |
|
|
.creator-label { |
|
|
color: #E0E0E0; |
|
|
font-size: 14px; |
|
|
margin-right: 8px; |
|
|
} |
|
|
.creator-link { |
|
|
display: flex; |
|
|
align-items: center; |
|
|
gap: 5px; |
|
|
color: #FF0033; |
|
|
text-decoration: none; |
|
|
transition: all 0.2s ease; |
|
|
} |
|
|
.creator-name { |
|
|
font-weight: 600; |
|
|
} |
|
|
.creator-arrow { |
|
|
font-size: 16px; |
|
|
transition: transform 0.2s ease; |
|
|
} |
|
|
.creator-link:hover .creator-arrow { |
|
|
transform: translateX(5px); |
|
|
} |
|
|
.link-arrow { |
|
|
display: inline-block; |
|
|
transition: transform 0.2s ease; |
|
|
} |
|
|
a:hover .link-arrow { |
|
|
transform: translateX(5px); |
|
|
} |
|
|
.axolotl-container { |
|
|
text-align: center; |
|
|
margin: 30px 0; |
|
|
filter: invert(1) hue-rotate(180deg); |
|
|
} |
|
|
.axolotl-container img { |
|
|
max-width: 300px; |
|
|
border-radius: 8px; |
|
|
box-shadow: 0 2px 10px rgba(0, 0, 0, 0.3); |
|
|
border: 1px solid #000000; |
|
|
} |
|
|
</style> |
|
|
<html lang="en"> |
|
|
<head> |
|
|
<meta charset="UTF-8"> |
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0"> |
|
|
<title>Rei-KTO-24B</title> |
|
|
</head> |
|
|
<body> |
|
|
<div class="container"> |
|
|
<div class="header"> |
|
|
<h1>Rei-KTO-24B</h1> |
|
|
</div> |
|
|
<div class="info"> |
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/66c26b6fb01b19d8c3c2467b/EFNIBpVv1OITZl_su5DyR.png" alt="Model banner"> |
|
|
<div style="text-align: center;"> |
|
|
<div class="creator-section"> |
|
|
<div class="creator-badge"> |
|
|
<span class="creator-label">Created by</span> |
|
|
<a href="https://huggingface.co/Delta-Vector" target="_blank" class="creator-link"> |
|
|
<span class="creator-name">Delta-Vector</span> |
|
|
<span class="creator-arrow">→</span> |
|
|
</a> |
|
|
</div> |
|
|
</div> |
|
|
<div class="model-info"> |
|
|
<h2>Model Information</h2> |
|
|
<div class="info-card"> |
|
|
<div class="info-header"> |
|
|
<h>Rei-KTO-24B</h3> |
|
|
<div class="model-tags"> |
|
|
<span class="model-tag">KTO enhanced</span> |
|
|
<span class="model-tag">Painted Fantasy Finetune</span> |
|
|
<span class="model-tag">Creative Prose</span> |
|
|
</div> |
|
|
</div> |
|
|
<div class="model-description"> |
|
|
<p>A model meant to replicate the style and prose of the Anthropic Claude models, Opus and Sonnet. This model is meant for Roleplaying/Creative-writing, Has some nice smarts without being too sloppy, etc - It's pretty good. Trained in 2 steps, Firstly SFT trained on Zerofata's PaintedFantasy which i found great at anime-otaku-esque characters, and then KTO'd to improve coherency and Instruct Following </p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
<div class="section"> |
|
|
<h2>Quantized Versions</h2> |
|
|
<div class="info-card"> |
|
|
<div class="model-composition"> |
|
|
<h4>Available Downloads</h4> |
|
|
<ul class="composition-list"> |
|
|
<li><span class="model-component"><a href="https://huggingface.co/mradermacher/Rei-24B-KTO-GGUF" target="_blank">GGUF Format</a></span>For use with LLama.cpp & Forks (Ty Mradermacher <3)</li> |
|
|
<li><span class="model-component"><a href="" target="_blank">EXL2 Format</a></span>For use with TabbyAPI (Coming Soon!)</li> |
|
|
</ul> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
<div class="section"> |
|
|
<h2>Prompting</h2> |
|
|
<p>The model is tuned with V7 Tekken formatting. A typical input would look like this:</p> |
|
|
<pre><code><span style="color:#FF0033;"><s>[SYSTEM_PROMPT]system_prompt[/SYSTEM_PROMPT][INST]Hi there![/INST]Nice to meet you!</s>[INST]Can I ask a question?[/INST] |
|
|
</code></pre> |
|
|
</div> |
|
|
<div class="section"> |
|
|
<h2>Training</h2> |
|
|
<p>Training was done in 2 steps, SFT>KTO</p> |
|
|
<div class="metrics-section"> |
|
|
<details> |
|
|
<summary>Access Configs</summary> |
|
|
<pre><code> SFT: https://wandb.ai/new-eden/Painted-Fantasy-Rei/artifacts/axolotl-config/config-u7to9d5q/v0/files/axolotl_config_f0p7vnaf.yml |
|
|
KTO : https://wandb.ai/new-eden/Painted-Rei/artifacts/axolotl-config/config-8n37w77c/v0/files/axolotl_config_hvrd2tzn.yml |
|
|
</code></pre> |
|
|
</details> |
|
|
</div> |
|
|
</div> |
|
|
<div class="section"> |
|
|
<h2>Training</h2> |
|
|
<p>The training was done for 2 epochs using 8 x A100s for 24 hours/p> |
|
|
<p style="text-align: center; margin-top: 20px;"> |
|
|
<div class="axolotl-container"> |
|
|
<a href="https://github.com/OpenAccess-AI-Collective/axolotl" target="_blank"> |
|
|
<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl"> |
|
|
</a> |
|
|
</div> |
|
|
<div class="section"> |
|
|
<h2>Credits</h2> |
|
|
<p>Thank you to <a href="https://huggingface.co/lucyknada">Lucy Knada</a>, <a href="https://huggingface.co/Ateron">Ateron</a>, <a href="https://huggingface.co/AliCat2">Alicat</a>, <a href="https://huggingface.co/intervitens">Intervitens</a>, <a href="https://huggingface.co/cgato">Cgato</a>, <a href="https://huggingface.co/kubernetes-bad">Kubernetes Bad</a> and the rest of <a href="https://huggingface.co/anthracite-org">Anthracite</a>.</p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</body> |
|
|
</html> |
