L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell by DavidAU

 ยป  All LLMs  ยป  DavidAU  ยป  L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell   URL Share it on

  Arxiv:2401.02415   Conversational   Creative   Creative writing   En   Extreme swearing   Fiction   Fiction story   Fiction writing   Graphic horror   Horror   Llama   Llama3   Mergekit   Not-for-all-audiences   Nsfw   Plot generation   Region:us   Roleplaying   Rp   Safetensors   Scene continue   Sharded   Story   Story generation   Storytelling   Sub-plot generation   Swearing   Tensorflow   Writing

L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell (DavidAU/L3-Grand-HORROR-18.5B-V1.8-STABLE-10-Gates-of-Hell)

L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell Parameters and Internals

Model Type 
text generation, creative writing, fiction writing, story generation
Use Cases 
Areas:
creative writing, fiction writing, storytelling, roleplaying, scene generation
Applications:
fiction writing, plot generation, sub-plot generation, scene continuation
Primary Use Cases:
general fiction, creative storytelling
Limitations:
horror bias, can generate extreme content, intended for creative tasks
Considerations:
Not suitable for all audiences due to uncensored content, including swearing and graphic horror.
Additional Notes 
STRONG horror bias with advanced prose capabilities, including dark humor and intense roleplay capabilities.
Supported Languages 
en (proficient)
Training Details 
Methodology:
Rebuilt from original L3-Stheno-Maid-Blackroot 8B using a new pass-through merge. Incorporated Brainstorm process.
Context Length:
8192
Input Output 
Input Format:
Designed for creative and detailed prompts.
Accepted Modalities:
text
Output Format:
Text responses with a range of dialogue, thoughts, and descriptive prose.
Performance Tips:
Use detailed prompts for richer scene generation. Model is capable of impromptu storytelling.
Release Notes 
Version:
V1.8
Notes:
Stabilized with new augmentation methods. Enhanced focus and prose quality. BRAINSTORM process integration for improved reasoning centers.
LLM NameL3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell
Repository ๐Ÿค—https://huggingface.co/DavidAU/L3-Grand-HORROR-18.5B-V1.8-STABLE-10-Gates-of-Hell 
Model Size18.5b
Required VRAM37.6 GB
Updated2025-04-22
MaintainerDavidAU
Model Typellama
Model Files  1.1 GB: 1-of-39   1.1 GB: 2-of-39   1.0 GB: 3-of-39   1.0 GB: 4-of-39   1.0 GB: 5-of-39   1.0 GB: 6-of-39   0.9 GB: 7-of-39   1.0 GB: 8-of-39   1.0 GB: 9-of-39   1.0 GB: 10-of-39   0.9 GB: 11-of-39   1.0 GB: 12-of-39   1.0 GB: 13-of-39   1.0 GB: 14-of-39   0.9 GB: 15-of-39   1.0 GB: 16-of-39   1.0 GB: 17-of-39   1.0 GB: 18-of-39   1.0 GB: 19-of-39   1.0 GB: 20-of-39   0.9 GB: 21-of-39   1.0 GB: 22-of-39   1.0 GB: 23-of-39   1.0 GB: 24-of-39   0.9 GB: 25-of-39   1.0 GB: 26-of-39   1.0 GB: 27-of-39   1.0 GB: 28-of-39   1.0 GB: 29-of-39   1.0 GB: 30-of-39   0.9 GB: 31-of-39   0.9 GB: 32-of-39   0.9 GB: 33-of-39   1.0 GB: 34-of-39   0.9 GB: 35-of-39   1.0 GB: 36-of-39   1.0 GB: 37-of-39   1.0 GB: 38-of-39   0.3 GB: 39-of-39
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length8192
Model Max Length8192
Transformers Version4.43.3
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typefloat16

Rank the L3 Grand HORROR 18.5B V1.8 STABLE 10 Gates Of Hell Capabilities

๐Ÿ†˜ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐ŸŒŸ

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 46565 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227