LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Dbrx Base Fixed by v2ray

 »  All LLMs  »  v2ray  »  Dbrx Base Fixed   URL Share it on

  15   
  Arxiv:2211.15841   Arxiv:2304.11277   Autotrain compatible   Conversational   Custom code   Dbrx   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/v2ray/dbrx-base-fixed 

Dbrx Base Fixed Benchmarks

LLME Score: 0.1757
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Dbrx Base Fixed (v2ray/dbrx-base-fixed)

Dbrx Base Fixed Parameters and Internals

Model Type 
transformer-based, decoder-only, large language model (LLM), mixture-of-experts (MoE) architecture
Use Cases 
Areas:
commercial applications, research applications
Applications:
general English-language tasks, coding tasks
Primary Use Cases:
text completion
Limitations:
not trained for non-English proficiency, generalist model
Considerations:
Should perform additional safety testing during fine-tuning for particular applications.
Supported Languages 
English (general proficiency)
Training Details 
Data Sources:
text, code
Data Volume:
12 trillion tokens
Methodology:
next-token prediction using a fine-grained mixture-of-experts (MoE) architecture with curriculum learning
Context Length:
32768
Model Architecture:
decoder-only large language model with rotary position encodings (RoPE), gated linear units (GLU), and grouped query attention (GQA)
Input Output 
Input Format:
text-based inputs
Accepted Modalities:
text
Output Format:
text-based outputs
Performance Tips:
FlashAttention2 can be used for faster inference if supported by GPU system.
Release Notes 
Version:
1.0
Notes:
Initial release of DBRX Base with fixed implementation for better training performance.
LLM NameDbrx Base Fixed
Repository 🤗https://huggingface.co/v2ray/dbrx-base-fixed 
Model Size131.6b
Required VRAM206.1 GB
Updated2025-02-22
Maintainerv2ray
Model Typedbrx
Model Files  5.0 GB: 1-of-54   4.9 GB: 2-of-54   4.9 GB: 3-of-54   4.9 GB: 4-of-54   4.9 GB: 5-of-54   4.9 GB: 6-of-54   4.9 GB: 7-of-54   4.9 GB: 8-of-54   4.9 GB: 9-of-54   4.9 GB: 10-of-54   4.9 GB: 11-of-54   4.9 GB: 12-of-54   4.9 GB: 13-of-54   4.9 GB: 14-of-54   4.9 GB: 15-of-54   4.9 GB: 16-of-54   4.9 GB: 17-of-54   4.9 GB: 18-of-54   4.9 GB: 19-of-54   4.9 GB: 20-of-54   4.9 GB: 21-of-54   4.9 GB: 22-of-54   4.9 GB: 23-of-54   4.9 GB: 24-of-54   4.9 GB: 25-of-54   4.9 GB: 26-of-54   4.9 GB: 27-of-54   4.9 GB: 28-of-54   4.9 GB: 29-of-54   4.9 GB: 30-of-54   4.9 GB: 31-of-54   4.9 GB: 32-of-54   4.9 GB: 33-of-54   4.9 GB: 34-of-54   4.9 GB: 35-of-54   5.0 GB: 36-of-54   5.0 GB: 37-of-54   4.9 GB: 38-of-54   4.9 GB: 39-of-54   4.9 GB: 40-of-54   4.9 GB: 41-of-54   4.9 GB: 42-of-54
Model ArchitectureDbrxForCausalLM
Licenseother
Transformers Version4.39.1
Vocabulary Size100352
Torch Data Typebfloat16

Best Alternatives to Dbrx Base Fixed

Best Alternatives
Context / RAM
Downloads
Likes
Dbrx Instruct0K / 181.1 GB132871110
...lphin 2.9.1 Dbrx Llamacppfixed0K / 186.6 GB82
Dbrx Base Converted V20K / 181.1 GB4536
Dolphin 2.9.1 Dbrx0K / 215.9 GB118
Dbrx Base0K / 185.5 GB19528
Dbrx Instruct0K / 181.1 GB2220
...rx Instruct Quantization Fixed0K / 176.9 GB159
Dbrx Base Quantization Fixed0K / 181.1 GB71
Note: green Score (e.g. "73.2") means that the model is better than v2ray/dbrx-base-fixed.

Rank the Dbrx Base Fixed Capabilities

🆘 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! 🌟

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? 43470 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer