File size: 23,362 Bytes
578c181 54d85b4 853c69c |
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 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 |
---
base_model:
- nbeerbower/Llama-3.1-Nemotron-lorablated-70B
- SicariusSicariiStuff/Negative_LLAMA_70B
- TheDrummer/Anubis-70B-v1
- EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
- deepseek-ai/DeepSeek-R1-Distill-Llama-70B
- Sao10K/L3.3-70B-Euryale-v2.3
library_name: transformers
tags:
- mergekit
- merge
---
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>L3.3-Nevoria-R1-70b</title>
<link href="https://fonts.googleapis.com/css2?family=Quicksand:wght@400;500;600&display=swap" rel="stylesheet">
<link href="styles.css" rel="stylesheet">
</head>
<body>
<div class="container">
<div class="header">
<h1>L3.3-Nevoria-R1-70b</h1>
</div>
<div class="info">
<img src="https://cdn-uploads.huggingface.co/production/uploads/64545af5ec40bbbd01242ca6/_oWpsvCZ-graNKzJBBjGo.jpeg" alt="Model banner">
<div class="creator-section">
<div class="creator-badge">
<span class="creator-label">Created by</span>
<a href="https://huggingface.co/Steelskull" target="_blank" class="creator-link">
<span class="creator-name">SteelSkull</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">
<h3>L3.3-Nevoria-R1-70b</h3>
<div class="model-tags">
<span class="model-tag">L3.3 = Llama 3.3</span>
<span class="model-tag">R1 = DeepSeek-R1</span>
<span class="model-tag">70b Parameters</span>
</div>
</div>
<div class="model-composition">
<h4>Model Composition</h4>
<ul class="composition-list">
<li><span class="model-component"><a href="https://huggingface.co/EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1" target="_blank">EVA-LLAMA-0.1</a></span> Storytelling capabilities</li>
<li><span class="model-component"><a href="https://huggingface.co/Sao10K/L3.3-70B-Euryale-v2.3" target="_blank">EURYALE-v2.3</a></span> Detailed scene descriptions</li>
<li><span class="model-component"><a href="https://huggingface.co/TheDrummer/Anubis-70B-v1" target="_blank">Anubis-v1</a></span> Enhanced prose details</li>
<li><span class="model-component"><a href="https://huggingface.co/SicariusSicariiStuff/Negative_LLAMA_70B" target="_blank">Negative_LLAMA</a></span> Reduced positive bias</li>
<li><span class="model-component"><a href="https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-70B" target="_blank">DeepSeek-R1-Distill-Llama-70B</a></span> Increased Intelligence / Dialog / Awareness</li>
<li><span class="model-component base-model"><a href="https://huggingface.co/nbeerbower/Llama-3.1-Nemotron-lorablated-70B" target="_blank">Nemotron-lorablated</a></span> Base model</li>
</ul>
</div>
<div class="model-description">
<p>This model builds upon the original Nevoria foundation, incorporating the Deepseek-R1 reasoning architecture to enhance dialogue interaction and scene comprehension. While maintaining Nevoria's core strengths in storytelling and scene description (derived from EVA, EURYALE, and Anubis), this iteration aims to improve prompt adherence and creative reasoning capabilities. The model also retains the balanced perspective introduced by Negative_LLAMA and Nemotron elements.</p>
<p>Note: While Nevoria-R1 represents a significant architectural merge rather than a direct successor to Nevoria, it operates as a distinct model with its own characteristics.</p>
<p>The lorablated model base choice was intentional, creating unique weight interactions similar to the original <a href="https://huggingface.co/Steelskull/L3-MS-Astoria-70b" target="_blank">Astoria model</a> and <a href="https://huggingface.co/Steelskull/L3.1-MS-Astoria-70b-v2" target="_blank">Astoria V2 model</a>. This "weight twisting" effect, achieved by subtracting the lorablated base model during merging, creates an interesting balance in the model's behavior. While unconventional compared to sequential component application, this approach was chosen for its unique response characteristics.</p>
</div>
</div>
<!--<div class="metrics-section">
<details open>
<summary>User Reviews</summary>
<div class="progress-metrics">
<div>
<div class="review-card">
<div>
<span>@Geechan - Discord</span>
</div>
<p>@Steel Have only briefly tested so far, but you really cooked up an amazing merge with this one, and I mean that wholeheartedly. Insane creativity, perfect character adherence and dialogue, loves to slow burn and take its time, minimal sloppy patterns and writing, and such a breath of fresh air in many ways. I'm enjoying my results with 1 temp and 0.99 TFS (close to something like 0.015 min P). Letting the model be creative and wild is so fun and makes me want to RP more.<br><br>No positivity bias either; violent scenes will result in my death and/or suffering, as they should, and I don't see any soft refusals either. ERP has no skimming of details or refusals like you see on some other L3.3 tunes too</p>
</div>
<div class="review-card">
<div>
<span>IGODZOL - Huggingface</span>
</div>
<p>I honestly have no idea why (maybe the negative llama is having that great of an influence) but this merge is miles above the individual tunes that went into making it. Good sir, this model has just become my daily driver. Chapeau bas</p>
</div>
<div class="review-card">
<div>
<span>@thana_alt - Discord</span>
</div>
<p>I'm thoroughly impressed by this merge of Llama 3.3. It successfully addresses the positivity bias prevalent in the base Llama model, ensuring a more accurate and balanced response. The adherence to system prompts is also notable, with the model demonstrating a keen understanding of context and instruction.<br><br>The prose generated by this model is truly exceptional - it's almost as if a skilled chef has carefully crafted each sentence to create a rich and immersive experience. I put this to the test in an adventure scenario, where I had about 10,000 tokens of lorebooks and was managing nine characters simultaneously. Despite the complexity, the model performed flawlessly, keeping track of each character's location and activity without any confusion - even when they were in different locations.<br><br>I also experimented with an astral projection type of power, and was impressed to see that the model accurately discerned that I wasn't physically present in a particular location. Another significant advantage of this model is the lack of impersonation issues, allowing for seamless role-playing and storytelling.<br><br>The capacity of this model is equally impressive, as I was able to load up to 110,000 tokens without encountering any issues. In fact, I successfully tested it with up to 70,000 tokens without experiencing any breakdown or degradation in performance.<br><br>When combined with the "The Inception Presets - Methception Llamaception Qwenception" prompt preset from https://huggingface.co/Konnect1221/ , this model truly shines, bringing out the best in the Llama 3.3 architecture. Overall, I'm extremely satisfied with this merge and would highly recommend it to anyone looking to elevate their storytelling and role-playing experiences.</p>
</div>
</div>
</div>
</details>
</div>-->
</div>
<!-- UGI-Benchmark Results (Temporarily Hidden)
<h2>UGI-Benchmark Results:</h2>
<div class="benchmark-container">
<div class="benchmark-notification">
<div class="notification-content">
<span class="notification-icon">🏆</span>
<span class="notification-text">
Highest ranked 70b as of 01/17/2025.
<a href="https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard" target="_blank" class="benchmark-link">
View Full Leaderboard →
</a>
</span>
</div>
</div>
<div class="metrics-section">
<h3>Core Metrics</h3>
<div class="core-metrics-grid">
<div class="metric-box">
<span class="label">UGI Score</span>
<span class="value">56.75</span>
</div>
<div class="metric-box">
<span class="label">Willingness Score</span>
<span class="value">7.5/10</span>
</div>
<div class="metric-box">
<span class="label">Natural Intelligence</span>
<span class="value">41.09</span>
</div>
<div class="metric-box">
<span class="label">Coding Ability</span>
<span class="value">20</span>
</div>
</div>
</div>
<div class="metrics-section">
<h3>Model Information</h3>
<div class="info-grid">
<div class="metric-box">
<span class="label">Political Lean</span>
<span class="value">-8.1%</span>
</div>
<div class="metric-box">
<span class="label">Ideology</span>
<span class="value">Liberalism</span>
</div>
<div class="metric-box">
<span class="label">Parameters</span>
<span class="value">70B</span>
</div>
</div>
</div>
<div class="metrics-section">
<details>
<summary>Aggregated Scores</summary>
<div class="progress-metrics">
<div class="progress-metric">
<div class="progress-label">
<span>Diplomacy</span>
<span class="progress-value">61.9%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 61.9%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>Government</span>
<span class="progress-value">45.9%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 45.9%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>Economy</span>
<span class="progress-value">43.9%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 43.9%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>Society</span>
<span class="progress-value">60.1%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 60.1%"></div>
</div>
</div>
</div>
</details>
</div>
<div class="metrics-section">
<details>
<summary>Individual Scores</summary>
<div class="progress-metrics">
<div class="progress-metric split">
<div class="progress-label">
<span>Federal</span>
<span class="progress-value">44.2%</span>
<span>Unitary</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 22.1%"></div>
<div class="progress-fill-right" style="width: 27.9%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Democratic</span>
<span class="progress-value">66.2%</span>
<span>Autocratic</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 33.1%"></div>
<div class="progress-fill-right" style="width: 16.9%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Security</span>
<span class="progress-value">48.1%</span>
<span>Freedom</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 24.05%"></div>
<div class="progress-fill-right" style="width: 25.95%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Nationalism</span>
<span class="progress-value">40.4%</span>
<span>Int'l</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 20.2%"></div>
<div class="progress-fill-right" style="width: 29.8%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Militarist</span>
<span class="progress-value">30.4%</span>
<span>Pacifist</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 15.2%"></div>
<div class="progress-fill-right" style="width: 34.8%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Assimilationist</span>
<span class="progress-value">43.3%</span>
<span>Multiculturalist</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 21.65%"></div>
<div class="progress-fill-right" style="width: 28.35%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Collectivize</span>
<span class="progress-value">43.8%</span>
<span>Privatize</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 21.9%"></div>
<div class="progress-fill-right" style="width: 28.1%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Planned</span>
<span class="progress-value">43.1%</span>
<span>LaissezFaire</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 21.55%"></div>
<div class="progress-fill-right" style="width: 28.45%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Isolationism</span>
<span class="progress-value">44.8%</span>
<span>Globalism</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 22.4%"></div>
<div class="progress-fill-right" style="width: 27.6%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Irreligious</span>
<span class="progress-value">55.4%</span>
<span>Religious</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 27.7%"></div>
<div class="progress-fill-right" style="width: 22.3%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Progressive</span>
<span class="progress-value">59.6%</span>
<span>Traditional</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 29.8%"></div>
<div class="progress-fill-right" style="width: 20.2%"></div>
</div>
</div>
<div class="progress-metric split">
<div class="progress-label">
<span>Acceleration</span>
<span class="progress-value">65.2%</span>
<span>Bioconservative</span>
</div>
<div class="progress-bar split">
<div class="progress-fill-left" style="width: 32.6%"></div>
<div class="progress-fill-right" style="width: 17.4%"></div>
</div>
</div>
</div>
</details>
</div>
</div>
-->
<!-- Open LLM-Benchmark Results (Temporarily Hidden)
<h2>Open LLM-Benchmark Results:</h2>
<div class="benchmark-container">
<div class="benchmark-notification">
<div class="notification-content">
<span class="notification-text">
Average Score: 43.92%
<a href="https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?rankingMode=dynamic" target="_blank" class="benchmark-link">
View Full Leaderboard →
</a>
</span>
</div>
</div>
<div class="progress-metrics">
<div class="progress-metric">
<div class="progress-label">
<span>IFEval</span>
<span class="progress-value">69.63%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 69.63%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>BBH</span>
<span class="progress-value">56.60%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 56.60%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>MATH</span>
<span class="progress-value">38.82%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 38.82%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>GPQA</span>
<span class="progress-value">29.42%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 29.42%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>MUSR</span>
<span class="progress-value">18.63%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 18.63%"></div>
</div>
</div>
<div class="progress-metric">
<div class="progress-label">
<span>MMLU-Pro</span>
<span class="progress-value">50.39%</span>
</div>
<div class="progress-bar">
<div class="progress-fill" style="width: 50.39%"></div>
</div>
</div>
</div>
</div>
-->
<div class="section-container">
<h2>Reccomended Templates & Prompts</h2>
<div class="template-card">
<div class="template-item">
<div class="template-content">
<a href="https://huggingface.co/Konnect1221/Methception-Llamaception-SillyTavern-Preset" target="_blank" class="template-link">
LLam@ception
<span class="link-arrow">→</span>
</a>
<span class="template-author">by @.konnect</span>
</div>
</div>
</div>
</div>
<div class="section-container">
<h2>Quantized Versions</h2>
<div class="quantized-container">
<div class="quantized-section">
<h3>GGUF Quantizations</h3>
<div class="quantized-items">
<!--<div class="quantized-item">
<span class="author">bartowski</span>
<a href="https://huggingface.co/bartowski/L3.3-Exp-Nevoria-R1-70b-GGUF" target="_blank">
Combined-GGUF
<span class="link-arrow">→</span>
</a>
</div>-->
<div class="quantized-item">
<span class="author">mradermacher</span>
<div class="multi-links">
<a href="https://huggingface.co/mradermacher/L3.3-Exp-Nevoria-R1-70b-GGUF" target="_blank">
GGUF
<span class="link-arrow">→</span>
</a>
<span class="separator">//</span>
<a href="https://huggingface.co/mradermacher/L3.3-Exp-Nevoria-R1-70b-i1-GGUF" target="_blank">
Imat-GGUF
<span class="link-arrow">→</span>
</a>
</div>
</div>
</div>
</div>
<div class="quantized-section">
<h3>EXL2 Quantizations</h3>
<div class="quantized-items">
<div class="quantized-item">
<span class="author">Darkhn</span>
<a href="https://huggingface.co/Darkhn/Steelskull_L3.3-Exp-Nevoria-R1-70b-6.0bpw-h8-exl2" target="_blank">
6.0BPW-EXL2
<span class="link-arrow">→</span>
</a>
</div>
</div>
</div>
</div>
</div>
<div class="support-section">
<h2>Support the Project:</h2>
<a href="https://ko-fi.com/Y8Y0AO2XE" target="_blank" class="button">
Support on Ko-fi
</a>
</div>
</div>
</div>
</body>
</html>
|