Update README.md
Browse files
README.md
CHANGED
|
@@ -1,34 +1,400 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
library_name: transformers
|
| 4 |
tags:
|
| 5 |
-
-
|
| 6 |
-
-
|
| 7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
-
# 24b
|
| 10 |
-
|
| 11 |
-
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
| 12 |
-
|
| 13 |
-
## Merge Details
|
| 14 |
-
### Merge Method
|
| 15 |
-
|
| 16 |
-
This model was merged using the Passthrough merge method using /home/quixi/storage/models/Austral-24B-Base + /home/quixi/storage/models/Austral-kto-lora as a base.
|
| 17 |
-
|
| 18 |
-
### Models Merged
|
| 19 |
-
|
| 20 |
-
The following models were included in the merge:
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
### Configuration
|
| 24 |
-
|
| 25 |
-
The following YAML configuration was used to produce this model:
|
| 26 |
-
|
| 27 |
-
```yaml
|
| 28 |
-
base_model: /home/quixi/storage/models/Austral-24B-Base+/home/quixi/storage/models/Austral-kto-lora
|
| 29 |
-
dtype: bfloat16
|
| 30 |
-
merge_method: passthrough
|
| 31 |
-
models:
|
| 32 |
-
- model: /home/quixi/storage/models/Austral-24B-Base+/home/quixi/storage/models/Austral-kto-lora
|
| 33 |
|
| 34 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model:
|
| 4 |
+
- LatitudeGames/Harbinger-24B
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
library_name: transformers
|
| 8 |
tags:
|
| 9 |
+
- roleplay
|
| 10 |
+
- finetune
|
| 11 |
+
- axolotl
|
| 12 |
+
- adventure
|
| 13 |
+
- creative-writing
|
| 14 |
+
- Mistral
|
| 15 |
+
- 24B
|
| 16 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
+
<<!DOCTYPE html>
|
| 19 |
+
<html lang="en">
|
| 20 |
+
<head>
|
| 21 |
+
<meta charset="UTF-8">
|
| 22 |
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
| 23 |
+
<title>Austral 24B Winton</title>
|
| 24 |
+
<link href="" rel="stylesheet">
|
| 25 |
+
<style>
|
| 26 |
+
body {
|
| 27 |
+
font-family: 'Roboto Slab', serif;
|
| 28 |
+
background: linear-gradient(135deg, #8B4513 0%, #A0522D 25%, #CD853F 50%, #D2691E 75%, #8B4513 100%);
|
| 29 |
+
background-size: 400% 400%;
|
| 30 |
+
animation: prehistoricShift 20s ease-in-out infinite;
|
| 31 |
+
color: #2F1B14;
|
| 32 |
+
margin: 0;
|
| 33 |
+
padding: 0;
|
| 34 |
+
font-size: 16px;
|
| 35 |
+
min-height: 100vh;
|
| 36 |
+
}
|
| 37 |
+
@keyframes prehistoricShift {
|
| 38 |
+
0%, 100% { background-position: 0% 50%; }
|
| 39 |
+
50% { background-position: 100% 50%; }
|
| 40 |
+
}
|
| 41 |
+
.container {
|
| 42 |
+
margin: 20px;
|
| 43 |
+
background: linear-gradient(145deg, #F4E4BC 0%, #DEB887 100%);
|
| 44 |
+
padding: 20px;
|
| 45 |
+
border-radius: 15px;
|
| 46 |
+
box-shadow: 0 8px 25px rgba(0, 0, 0, 0.4), inset 0 2px 5px rgba(255, 255, 255, 0.3);
|
| 47 |
+
border: 4px solid #8B4513;
|
| 48 |
+
position: relative;
|
| 49 |
+
overflow: hidden;
|
| 50 |
+
}
|
| 51 |
+
.container::before {
|
| 52 |
+
content: '';
|
| 53 |
+
position: absolute;
|
| 54 |
+
top: 0;
|
| 55 |
+
left: 0;
|
| 56 |
+
right: 0;
|
| 57 |
+
bottom: 0;
|
| 58 |
+
background-image:
|
| 59 |
+
radial-gradient(circle at 20% 80%, rgba(139, 69, 19, 0.1) 0%, transparent 50%),
|
| 60 |
+
radial-gradient(circle at 80% 20%, rgba(160, 82, 45, 0.1) 0%, transparent 50%);
|
| 61 |
+
pointer-events: none;
|
| 62 |
+
}
|
| 63 |
+
.header h1 {
|
| 64 |
+
font-family: 'Cinzel', serif;
|
| 65 |
+
font-size: 32px;
|
| 66 |
+
color: #5D2E0C;
|
| 67 |
+
margin: 0 0 20px 0;
|
| 68 |
+
text-align: center;
|
| 69 |
+
text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);
|
| 70 |
+
letter-spacing: 2px;
|
| 71 |
+
position: relative;
|
| 72 |
+
}
|
| 73 |
+
.section {
|
| 74 |
+
margin-top: 30px;
|
| 75 |
+
position: relative;
|
| 76 |
+
}
|
| 77 |
+
.section h2 {
|
| 78 |
+
font-family: 'Cinzel', serif;
|
| 79 |
+
font-size: 26px;
|
| 80 |
+
color: #5D2E0C;
|
| 81 |
+
text-align: center;
|
| 82 |
+
margin-bottom: 20px;
|
| 83 |
+
text-shadow: 1px 1px 2px rgba(0, 0, 0, 0.2);
|
| 84 |
+
letter-spacing: 1px;
|
| 85 |
+
}
|
| 86 |
+
.info p {
|
| 87 |
+
color: #2F1B14;
|
| 88 |
+
line-height: 1.7;
|
| 89 |
+
font-size: 16px;
|
| 90 |
+
text-shadow: 0 1px 1px rgba(255, 255, 255, 0.5);
|
| 91 |
+
}
|
| 92 |
+
.info img {
|
| 93 |
+
width: 85%;
|
| 94 |
+
border-radius: 12px;
|
| 95 |
+
margin: 0 auto 15px;
|
| 96 |
+
display: block;
|
| 97 |
+
box-shadow: 0 0 25px rgba(0, 0, 0, 0.4);
|
| 98 |
+
border: 3px solid #8B4513;
|
| 99 |
+
filter: sepia(20%) contrast(110%);
|
| 100 |
+
}
|
| 101 |
+
a {
|
| 102 |
+
color: #5D2E0C;
|
| 103 |
+
text-decoration: none;
|
| 104 |
+
transition: all 0.3s ease;
|
| 105 |
+
font-weight: 500;
|
| 106 |
+
}
|
| 107 |
+
a:hover {
|
| 108 |
+
color: #8B4513;
|
| 109 |
+
text-shadow: 1px 1px 2px rgba(0, 0, 0, 0.2);
|
| 110 |
+
}
|
| 111 |
+
.button {
|
| 112 |
+
display: inline-block;
|
| 113 |
+
background: linear-gradient(145deg, #CD853F, #D2691E);
|
| 114 |
+
color: #2F1B14;
|
| 115 |
+
padding: 12px 24px;
|
| 116 |
+
border-radius: 8px;
|
| 117 |
+
cursor: pointer;
|
| 118 |
+
text-decoration: none;
|
| 119 |
+
transition: all 0.3s ease;
|
| 120 |
+
box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
|
| 121 |
+
border: 2px solid #8B4513;
|
| 122 |
+
}
|
| 123 |
+
.button:hover {
|
| 124 |
+
background: linear-gradient(145deg, #D2691E, #CD853F);
|
| 125 |
+
box-shadow: 0 6px 15px rgba(139, 69, 19, 0.4);
|
| 126 |
+
transform: translateY(-2px);
|
| 127 |
+
}
|
| 128 |
+
pre {
|
| 129 |
+
background: linear-gradient(145deg, #F5DEB3, #DEB887);
|
| 130 |
+
padding: 20px;
|
| 131 |
+
border-radius: 8px;
|
| 132 |
+
overflow-x: auto;
|
| 133 |
+
border: 2px solid #8B4513;
|
| 134 |
+
box-shadow: inset 0 2px 5px rgba(0, 0, 0, 0.1);
|
| 135 |
+
}
|
| 136 |
+
code {
|
| 137 |
+
font-family: 'Courier New', monospace;
|
| 138 |
+
color: #2F1B14;
|
| 139 |
+
}
|
| 140 |
+
.info-card {
|
| 141 |
+
background: linear-gradient(145deg, #F5DEB3, #DEB887);
|
| 142 |
+
border: 3px solid #8B4513;
|
| 143 |
+
border-radius: 12px;
|
| 144 |
+
overflow: hidden;
|
| 145 |
+
box-shadow: 0 6px 15px rgba(0, 0, 0, 0.2);
|
| 146 |
+
}
|
| 147 |
+
.info-header {
|
| 148 |
+
background: linear-gradient(145deg, #CD853F, #D2691E);
|
| 149 |
+
padding: 25px;
|
| 150 |
+
border-bottom: 2px solid #8B4513;
|
| 151 |
+
}
|
| 152 |
+
.info-header h3 {
|
| 153 |
+
font-family: 'Cinzel', serif;
|
| 154 |
+
color: #2F1B14;
|
| 155 |
+
margin: 0 0 15px 0;
|
| 156 |
+
font-size: 22px;
|
| 157 |
+
text-align: center;
|
| 158 |
+
text-shadow: 1px 1px 2px rgba(0, 0, 0, 0.2);
|
| 159 |
+
letter-spacing: 1px;
|
| 160 |
+
}
|
| 161 |
+
.model-tags {
|
| 162 |
+
display: flex;
|
| 163 |
+
gap: 10px;
|
| 164 |
+
flex-wrap: wrap;
|
| 165 |
+
justify-content: center;
|
| 166 |
+
}
|
| 167 |
+
.model-tag {
|
| 168 |
+
background: linear-gradient(145deg, #DEB887, #CD853F);
|
| 169 |
+
color: #2F1B14;
|
| 170 |
+
padding: 6px 12px;
|
| 171 |
+
border-radius: 6px;
|
| 172 |
+
font-size: 12px;
|
| 173 |
+
border: 2px solid #8B4513;
|
| 174 |
+
font-weight: 500;
|
| 175 |
+
box-shadow: 0 2px 4px rgba(0, 0, 0, 0.1);
|
| 176 |
+
}
|
| 177 |
+
.model-composition {
|
| 178 |
+
padding: 25px;
|
| 179 |
+
border-bottom: 2px solid #8B4513;
|
| 180 |
+
}
|
| 181 |
+
.model-composition h4 {
|
| 182 |
+
font-family: 'Cinzel', serif;
|
| 183 |
+
color: #5D2E0C;
|
| 184 |
+
margin: 0 0 20px 0;
|
| 185 |
+
font-size: 18px;
|
| 186 |
+
text-align: center;
|
| 187 |
+
letter-spacing: 1px;
|
| 188 |
+
}
|
| 189 |
+
.composition-list {
|
| 190 |
+
list-style: none;
|
| 191 |
+
padding: 0;
|
| 192 |
+
margin: 0;
|
| 193 |
+
display: grid;
|
| 194 |
+
gap: 15px;
|
| 195 |
+
}
|
| 196 |
+
.composition-list li {
|
| 197 |
+
color: #2F1B14;
|
| 198 |
+
display: flex;
|
| 199 |
+
align-items: baseline;
|
| 200 |
+
gap: 12px;
|
| 201 |
+
padding: 10px;
|
| 202 |
+
background: rgba(245, 222, 179, 0.5);
|
| 203 |
+
border-radius: 6px;
|
| 204 |
+
border-left: 4px solid #8B4513;
|
| 205 |
+
}
|
| 206 |
+
.model-component {
|
| 207 |
+
font-weight: 600;
|
| 208 |
+
min-width: 120px;
|
| 209 |
+
}
|
| 210 |
+
.model-description {
|
| 211 |
+
padding: 25px;
|
| 212 |
+
background: linear-gradient(145deg, #F5DEB3, #F4E4BC);
|
| 213 |
+
}
|
| 214 |
+
.metrics-section {
|
| 215 |
+
margin-bottom: 30px;
|
| 216 |
+
}
|
| 217 |
+
.metrics-section details {
|
| 218 |
+
background: linear-gradient(145deg, #F5DEB3, #DEB887);
|
| 219 |
+
border: 3px solid #8B4513;
|
| 220 |
+
border-radius: 10px;
|
| 221 |
+
padding: 20px;
|
| 222 |
+
margin-bottom: 20px;
|
| 223 |
+
box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
|
| 224 |
+
}
|
| 225 |
+
.metrics-section summary {
|
| 226 |
+
font-family: 'Cinzel', serif;
|
| 227 |
+
color: #5D2E0C;
|
| 228 |
+
font-size: 18px;
|
| 229 |
+
cursor: pointer;
|
| 230 |
+
outline: none;
|
| 231 |
+
padding: 10px 0;
|
| 232 |
+
text-align: center;
|
| 233 |
+
font-weight: 500;
|
| 234 |
+
letter-spacing: 1px;
|
| 235 |
+
}
|
| 236 |
+
.creator-section {
|
| 237 |
+
margin: 25px 0;
|
| 238 |
+
text-align: center;
|
| 239 |
+
}
|
| 240 |
+
.creator-badge {
|
| 241 |
+
display: inline-flex;
|
| 242 |
+
align-items: center;
|
| 243 |
+
background: linear-gradient(145deg, #CD853F, #D2691E);
|
| 244 |
+
border: 3px solid #8B4513;
|
| 245 |
+
border-radius: 10px;
|
| 246 |
+
padding: 15px 20px;
|
| 247 |
+
box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
|
| 248 |
+
}
|
| 249 |
+
.creator-label {
|
| 250 |
+
color: #2F1B14;
|
| 251 |
+
font-size: 14px;
|
| 252 |
+
margin-right: 10px;
|
| 253 |
+
font-weight: 500;
|
| 254 |
+
}
|
| 255 |
+
.creator-link {
|
| 256 |
+
display: flex;
|
| 257 |
+
align-items: center;
|
| 258 |
+
gap: 8px;
|
| 259 |
+
color: #2F1B14;
|
| 260 |
+
text-decoration: none;
|
| 261 |
+
transition: all 0.3s ease;
|
| 262 |
+
}
|
| 263 |
+
.creator-name {
|
| 264 |
+
font-weight: 600;
|
| 265 |
+
}
|
| 266 |
+
.creator-arrow {
|
| 267 |
+
font-size: 16px;
|
| 268 |
+
transition: transform 0.3s ease;
|
| 269 |
+
}
|
| 270 |
+
.creator-link:hover .creator-arrow {
|
| 271 |
+
transform: translateX(5px);
|
| 272 |
+
}
|
| 273 |
+
.link-arrow {
|
| 274 |
+
display: inline-block;
|
| 275 |
+
transition: transform 0.3s ease;
|
| 276 |
+
}
|
| 277 |
+
a:hover .link-arrow {
|
| 278 |
+
transform: translateX(3px);
|
| 279 |
+
}
|
| 280 |
+
.axolotl-container {
|
| 281 |
+
text-align: center;
|
| 282 |
+
margin: 35px 0;
|
| 283 |
+
}
|
| 284 |
+
.axolotl-container img {
|
| 285 |
+
max-width: 300px;
|
| 286 |
+
border-radius: 10px;
|
| 287 |
+
box-shadow: 0 6px 15px rgba(0, 0, 0, 0.3);
|
| 288 |
+
border: 3px solid #8B4513;
|
| 289 |
+
filter: sepia(30%) contrast(110%);
|
| 290 |
+
}
|
| 291 |
+
/* Fossil texture overlay */
|
| 292 |
+
.fossil-texture {
|
| 293 |
+
position: relative;
|
| 294 |
+
}
|
| 295 |
+
.fossil-texture::after {
|
| 296 |
+
content: '';
|
| 297 |
+
position: absolute;
|
| 298 |
+
top: 0;
|
| 299 |
+
left: 0;
|
| 300 |
+
right: 0;
|
| 301 |
+
bottom: 0;
|
| 302 |
+
background-image:
|
| 303 |
+
radial-gradient(circle at 25% 25%, rgba(139, 69, 19, 0.05) 2px, transparent 2px),
|
| 304 |
+
radial-gradient(circle at 75% 75%, rgba(160, 82, 45, 0.05) 1px, transparent 1px);
|
| 305 |
+
background-size: 50px 50px, 30px 30px;
|
| 306 |
+
pointer-events: none;
|
| 307 |
+
}
|
| 308 |
+
</style>
|
| 309 |
+
</head>
|
| 310 |
+
<body>
|
| 311 |
+
<div class="container fossil-texture">
|
| 312 |
+
<div class="header">
|
| 313 |
+
<h1>Austral 24B Winton</h1>
|
| 314 |
+
</p>
|
| 315 |
+
</div>
|
| 316 |
+
<div class="info">
|
| 317 |
+
<img src="" alt="Model banner">
|
| 318 |
+
<div style="text-align: center;">
|
| 319 |
+
<div class="creator-section">
|
| 320 |
+
<div class="creator-badge">
|
| 321 |
+
<span class="creator-label">Trained by</span>
|
| 322 |
+
<a href="https://huggingface.co/Delta-Vector" target="_blank" class="creator-link">
|
| 323 |
+
<span class="creator-name">Delta-Vector</span>
|
| 324 |
+
</a>
|
| 325 |
+
</div>
|
| 326 |
+
</div>
|
| 327 |
+
<div class="model-info">
|
| 328 |
+
<h2>Overview</h2>
|
| 329 |
+
<div class="info-card">
|
| 330 |
+
<div class="info-header">
|
| 331 |
+
<h3>Austral 24B - Winton</h3>
|
| 332 |
+
<div class="model-tags">
|
| 333 |
+
<span class="model-tag">Harbinger Finetune</span>
|
| 334 |
+
<span class="model-tag">Mistral Based</span>
|
| 335 |
+
<span class ="model-tag">KTO enhanced</span>
|
| 336 |
+
<span class ="model-tag">Adventure/Roleplay generalist</span>
|
| 337 |
+
<span class="model-tag">24B Sized model</span>
|
| 338 |
+
</div>
|
| 339 |
+
</div>
|
| 340 |
+
<div class="model-description">
|
| 341 |
+
<p style="font-weight: bold; font-style: italic;">More than 1.5-metres tall, about six-metres long and up to 1000-kilograms heavy, Australovenator Wintonensis was a fast and agile hunter. The largest known Australian theropod.</p>
|
| 342 |
+
<p>This is a finetune of Harbinger 24B to be a generalist Roleplay/Adventure model. I've removed some of the "slops" that i noticed in an otherwise great model aswell as improving the general writing of the model, This was a multi-stage finetune, all previous checkpoints are released aswell.</p>
|
| 343 |
+
<p>Support my finetunes / Me on Kofi: https://Ko-fi.com/deltavector | Thank you to Auri for helping/Testing ♥</p>
|
| 344 |
+
</div>
|
| 345 |
+
</div>
|
| 346 |
+
</div>
|
| 347 |
+
<div class="section">
|
| 348 |
+
<h2>Quants</h2>
|
| 349 |
+
<div class="info-card">
|
| 350 |
+
<div class="model-composition">
|
| 351 |
+
<h4>Quants Formats</h4>
|
| 352 |
+
<ul class="composition-list">
|
| 353 |
+
<li><span class="model-component"><a href="" target="_blank">GGUF</a></span>For use with LLama.cpp & Forks (Soon to be made!)</li>
|
| 354 |
+
<li><span class="model-component"><a href="" target="_blank">EXL3</a></span>For use with TabbyAPI (coming soon!)</li>
|
| 355 |
+
<li><span class="model-component"><a href="" target="_blank">FP8</a></span>For use with Aphrodite/VLLM</li>
|
| 356 |
+
</ul>
|
| 357 |
+
</div>
|
| 358 |
+
</div>
|
| 359 |
+
</div>
|
| 360 |
+
<div class="section">
|
| 361 |
+
<h2>Chat Format</h2>
|
| 362 |
+
<p>This model utilizes ChatML.</p>
|
| 363 |
+
<pre><code><|im_start|>user
|
| 364 |
+
Hi there!<|im_end|>
|
| 365 |
+
<|im_start|>assistant
|
| 366 |
+
Nice to meet you!<|im_end|>
|
| 367 |
+
<|im_start|>user
|
| 368 |
+
Can I ask a question?<|im_end|>
|
| 369 |
+
<|im_start|>assistant</code></pre>
|
| 370 |
+
</div>
|
| 371 |
+
<div class="section">
|
| 372 |
+
<h2>Training</h2>
|
| 373 |
+
<p>As the the Austral/Francois tradition, I built off another great finetune Harbinger-24B, I did 4 epochs ontop with roughly the same datamix as Francois-Huali/Austral 70B as a R128 Lora, then KTO alignment with a mix of Instruct/Small writing datasets and then finally another 4 epoch SFT with Rep_remover (Thanks Pocket!)</p>
|
| 374 |
+
<div class="metrics-section">
|
| 375 |
+
<details>
|
| 376 |
+
<summary>Config(Post-KTO SFT)</summary>
|
| 377 |
+
<pre><code>https://wandb.ai/new-eden/austral/artifacts/axolotl-config/config-0tzehrhe/v0/files/axolotl_config_m8018fm4.yml</code></pre>
|
| 378 |
+
</details>
|
| 379 |
+
</div>
|
| 380 |
+
</div>
|
| 381 |
+
<div class="section">
|
| 382 |
+
<h2Fossil Formation Process</h2>
|
| 383 |
+
<p>This model was trained over 4 epochs using 8 x A100s for the base SFT, Then i used KTO to clean up some coherency issues for 1 epoch, then finally training for another 4 epochs on Rep_Remover to delete slops. Total was roughly 80 hours total.</p>
|
| 384 |
+
<p style="text-align: center; margin-top: 20px;">
|
| 385 |
+
<div class="axolotl-container">
|
| 386 |
+
<a href="https://github.com/OpenAccess-AI-Collective/axolotl" target="_blank">
|
| 387 |
+
<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl">
|
| 388 |
+
</a>
|
| 389 |
+
</div>
|
| 390 |
+
<div class="section">
|
| 391 |
+
<h2>Credits</h2>
|
| 392 |
+
<p>TYSM to my friends: Auri, Lucy, Trappu, Alicat, Kubernetes Bad, Intervitens, NyxKrage & Kalomaze</p>
|
| 393 |
+
</p>
|
| 394 |
+
</div>
|
| 395 |
+
</div>
|
| 396 |
+
</div>
|
| 397 |
+
</div>
|
| 398 |
+
</div>
|
| 399 |
+
</body>
|
| 400 |
+
</html>
|