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

MoMo 72B Lora 1.8.7 DPO by moreh

 »  All LLMs  »  moreh  »  MoMo 72B Lora 1.8.7 DPO   URL Share it on

  21   
  Arxiv:2106.09685   Arxiv:2305.18290   Autotrain compatible   En   Endpoints compatible   Llama   Lora   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/moreh/MoMo-72B-lora-1.8.7-DPO 

MoMo 72B Lora 1.8.7 DPO Benchmarks

MMLU Pro: 40.26
GPQA: 9.84
MUSR: 14.42
BBH: 43.13
IFEval: 51.67 vs 88 (so35)-41.3%
ARC: 70.82 vs 96.7 (so35)-26.8%
HellaSwag: 85.96 vs 95.3 (gpt4)-9.8%
MMLU: 77.13 vs 88.3 (so35)-12.7%
TruthfulQA: 74.71 vs 59 (gpt4)26.6%
WinoGrande: 84.06 vs 87.5 (gpt4)-3.9%
GSM8K: 78.62 vs 96.4 (so35)-18.4%
MATH Lvl 5: 18.13
LLME Score: 0.23724
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MoMo 72B Lora 1.8.7 DPO (moreh/MoMo-72B-lora-1.8.7-DPO)

MoMo 72B Lora 1.8.7 DPO Parameters and Internals

Additional Notes 
For more information on the MoAI platform and its capabilities, refer to https://moreh.io/product.
Supported Languages 
languages (en), proficiency ()
Training Details 
Data Sources:
Open-Orca/SlimOrca, https://huggingface.co/datasets/jondurbin/truthy-dpo-v0.1, https://huggingface.co/datasets/Intel/orca_dpo_pairs
Methodology:
Direct Preference Optimization (DPO), Supervised Fine-Tuning (SFT) using LoRA.
Hardware Used:
AMD MI250
Release Notes 
Version:
V1.8.7
Date:
24/04/05
Notes:
Model was trained with optimizations in hyperparameters without exploiting weight merge. Adjustments for compatibility with Llama for leaderboard submission.
LLM NameMoMo 72B Lora 1.8.7 DPO
Repository 🤗https://huggingface.co/moreh/MoMo-72B-lora-1.8.7-DPO 
Model Size72b
Required VRAM208.5 GB
Updated2025-05-31
Maintainermoreh
Model Typellama
Model Files  5.0 GB: 1-of-63   4.6 GB: 2-of-63   4.3 GB: 3-of-63   4.3 GB: 4-of-63   4.8 GB: 5-of-63   4.8 GB: 6-of-63   4.3 GB: 7-of-63   4.8 GB: 8-of-63   4.8 GB: 9-of-63   4.3 GB: 10-of-63   4.8 GB: 11-of-63   4.8 GB: 12-of-63   4.3 GB: 13-of-63   4.8 GB: 14-of-63   4.8 GB: 15-of-63   4.3 GB: 16-of-63   4.8 GB: 17-of-63   4.8 GB: 18-of-63   4.3 GB: 19-of-63   4.8 GB: 20-of-63   4.8 GB: 21-of-63   4.3 GB: 22-of-63   4.8 GB: 23-of-63   4.8 GB: 24-of-63   4.3 GB: 25-of-63   4.8 GB: 26-of-63   4.8 GB: 27-of-63   4.3 GB: 28-of-63   4.8 GB: 29-of-63   4.8 GB: 30-of-63   4.3 GB: 31-of-63   4.8 GB: 32-of-63   4.8 GB: 33-of-63   4.3 GB: 34-of-63   4.8 GB: 35-of-63   4.8 GB: 36-of-63   4.3 GB: 37-of-63   4.8 GB: 38-of-63   4.8 GB: 39-of-63   4.3 GB: 40-of-63   4.8 GB: 41-of-63   4.8 GB: 42-of-63   4.3 GB: 43-of-63   4.8 GB: 44-of-63   4.8 GB: 45-of-63
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensemit
Context Length32768
Model Max Length32768
Transformers Version4.36.0
Vocabulary Size152064
LoRA ModelYes
Torch Data Typefloat32

Quantized Models of the MoMo 72B Lora 1.8.7 DPO

Model
Likes
Downloads
VRAM
Smaug 72B V0.1 2.4bpw H6 EXL211224 GB
Smaug 72B V0.1 GPTQ81741 GB
Smaug 72B V0.1 AWQ91141 GB
MoMo 72B Lora 1.8.7 DPO GPTQ72441 GB

Best Alternatives to MoMo 72B Lora 1.8.7 DPO

Best Alternatives
Context / RAM
Downloads
Likes
2 Pro Math128K / 141.9 GB70
TW3 JRGL V232K / 79.7 GB64250
Smaug 72B V0.132K / 144.5 GB3730469
Le Triomphant ECE TW332K / 79.7 GB64354
ECE TW3 JRGL V532K / 159.6 GB32830
Rhea 72B V0.532K / 144.5 GB3314136
JuliusCesar 72B BeyonderV.032K / 74.2 GB170
Caigun Model 72B KGI32K / 144.6 GB110
Rhea 125 V0.532K / 249 GB231
Exodius 70B32K / 144.6 GB100
Note: green Score (e.g. "73.2") means that the model is better than moreh/MoMo-72B-lora-1.8.7-DPO.

Rank the MoMo 72B Lora 1.8.7 DPO 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? 47753 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