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

Mpt 30B Instruct by mosaicml

 »  All LLMs  »  mosaicml  »  Mpt 30B Instruct   URL Share it on

  1263   
  Arxiv:2108.12409   Arxiv:2205.14135   Autotrain compatible   Composer   Custom code   Dataset:allenai/qasper   Dataset:competition math   Dataset:duorc   Dataset:emozilla/quality   Dataset:gsm8k   Dataset:knkarthick/dialogsum   Dataset:mosaicml/dolly hhrlhf   Dataset:scrolls/summ screen fd   Dataset:spider   Instruct   Llm-foundry   Mosaicml   Mpt   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/mosaicml/mpt-30b-instruct 

Mpt 30B Instruct Benchmarks

ARC: 58.45 vs 96.7 (so35)-39.6%
HellaSwag: 84.31 vs 95.3 (gpt4)-11.5%
MMLU: 49.15 vs 88.3 (so35)-44.3%
TruthfulQA: 38.05 vs 59 (gpt4)-35.5%
WinoGrande: 75.14 vs 87.5 (gpt4)-14.1%
GSM8K: 15.31 vs 96.4 (so35)-84.1%
LLME Score: 0.17014
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Mpt 30B Instruct (mosaicml/mpt-30b-instruct)

Mpt 30B Instruct Parameters and Internals

Model Type 
text generation, instruction following
Use Cases 
Areas:
Research, Commercial applications
Primary Use Cases:
Short-form instruction following tasks
Limitations:
May produce factually incorrect output and offensive content
Training Details 
Data Sources:
Dolly HHRLHF derived from Databricks Dolly-15k, Anthropic Helpful and Harmless (HH-RLHF), Competition Math, Duorc, CoT GSM8k, Qasper, Quality, Summ Screen FD, Spider
Methodology:
The model was finetuned on short-form instruction following tasks from various datasets.
Context Length:
8192
Training Time:
8 hours on 72 A100 40GB GPUs
Hardware Used:
72 A100 40GB GPUs
Model Architecture:
A modification of a standard decoder-only transformer, utilizes FlashAttention and ALiBi with no positional embeddings.
Input Output 
Input Format:
Prompts formatted with instructions preceding responses.
Accepted Modalities:
text
Output Format:
Generated text responses
Performance Tips:
Use torch.autocast for lower precision performance, and `trust_remote_code=True` to load custom architecture.
LLM NameMpt 30B Instruct
Repository 🤗https://huggingface.co/mosaicml/mpt-30b-instruct 
Model Size30b
Required VRAM60.1 GB
Updated2024-12-22
Maintainermosaicml
Model Typempt
Instruction-BasedYes
Model Files  9.8 GB: 1-of-7   9.9 GB: 2-of-7   9.9 GB: 3-of-7   9.9 GB: 4-of-7   9.9 GB: 5-of-7   9.9 GB: 6-of-7   0.8 GB: 7-of-7
Model ArchitectureMPTForCausalLM
Licenseapache-2.0
Model Max Length8192
Transformers Version4.28.1
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50432
Torch Data Typebfloat16

Quantized Models of the Mpt 30B Instruct

Model
Likes
Downloads
VRAM
Mpt 30B Instruct Q851630 GB

Best Alternatives to Mpt 30B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
...t 30B Instruct Peft Compatible0K / 60.1 GB132
Ct2fast Mpt 30B Instruct0K / 30 GB94
Ct2fast Mpt 30B Chat0K / 30 GB102
Mpt 30B Chat Q80K / 30.4 GB191
Mpt 30B Instruct Q80K / 30.4 GB165
...l Mpt 30B Instruct W4 G128 AWQ0K / 16.1 GB82
Note: green Score (e.g. "73.2") means that the model is better than mosaicml/mpt-30b-instruct.

Rank the Mpt 30B Instruct 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? 40066 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 v20241217
  Support LLM Explorer