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

Dolphin 2.6 Mistral 7B DPO by cognitivecomputations

 »  All LLMs  »  cognitivecomputations  »  Dolphin 2.6 Mistral 7B DPO   URL Share it on

  2228   
  Autotrain compatible Dataset:argilla/ultrafeedback-...   Dataset:ehartford/dolphin Dataset:ehartford/dolphin-code... Dataset:ise-uiuc/magicoder-evo... Dataset:ise-uiuc/magicoder-oss... Dataset:jondurbin/airoboros-2....   Dataset:ldjnr/capybara   Dataset:teknium/openhermes   En   Endpoints compatible   Instruct   Mistral   Model-index   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b-dpo 

Dolphin 2.6 Mistral 7B DPO Benchmarks

ARC: 65.61 vs 96.7 (so35)-32.2%
HellaSwag: 85.48 vs 95.3 (gpt4)-10.3%
MMLU: 63.24 vs 88.3 (so35)-28.4%
TruthfulQA: 61.47 vs 59 (gpt4)4.2%
WinoGrande: 78.61 vs 87.5 (gpt4)-10.2%
GSM8K: 48.75 vs 96.4 (so35)-49.4%
LLME Score: 0.19758
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Dolphin 2.6 Mistral 7B DPO (cognitivecomputations/dolphin-2.6-mistral-7b-dpo)

Dolphin 2.6 Mistral 7B DPO Parameters and Internals

Model Type 
text-generation
Additional Notes 
Implement own alignment layer before service deployment to handle compliance.
Training Details 
Data Sources:
ehartford/dolphin, jondurbin/airoboros-2.2.1, ehartford/dolphin-coder, teknium/openhermes, ise-uiuc/Magicoder-OSS-Instruct-75K, ise-uiuc/Magicoder-Evol-Instruct-110K, LDJnr/Capybara, argilla/ultrafeedback-binarized-preferences-cleaned
Methodology:
DPO tuned
Context Length:
16000
Training Time:
2 days
Hardware Used:
4x A100s
Model Architecture:
Mistral-7b
Input Output 
Input Format:
Uses ChatML prompt format.
Accepted Modalities:
text
Performance Tips:
Encourage compliance in system prompt.
Release Notes 
Version:
2.6 DPO
Notes:
DPO tuned on argilla/ultrafeedback-binarized-preferences-cleaned; uncensored; filtered dataset to remove alignment and bias.
LLM NameDolphin 2.6 Mistral 7B DPO
Repository 🤗https://huggingface.co/cognitivecomputations/dolphin-2.6-mistral-7b-dpo 
Model Size7b
Required VRAM14.4 GB
Updated2025-02-22
Maintainercognitivecomputations
Model Typemistral
Instruction-BasedYes
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<|im_end|>
Vocabulary Size32001
Torch Data Typebfloat16

Quantized Models of the Dolphin 2.6 Mistral 7B DPO

Model
Likes
Downloads
VRAM
...olphin 2.6 Mistral 7B DPO GGUF214693 GB
Dolphin 2.6 Mistral 7B DPO AWQ2744 GB
...olphin 2.6 Mistral 7B DPO GPTQ9404 GB

Best Alternatives to Dolphin 2.6 Mistral 7B DPO

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB462011
SpydazWeb AI HumanAI RP512K / 14.4 GB121
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
... Summarize 64K QLoRANET Merged128K / 4.1 GB120
...1 Summarize 64K LoRANET Merged128K / 14.4 GB110
Mistral 7B Instruct V0.232K / 14.4 GB33160952630
Mistral 7B Instruct V0.132K / 14.4 GB1916301573
...ity Instruct 7M Gen Mistral 7B32K / 14.4 GB37505
...ty Instruct 3M 0625 Mistral 7B32K / 14.4 GB37063
Note: green Score (e.g. "73.2") means that the model is better than cognitivecomputations/dolphin-2.6-mistral-7b-dpo.

Rank the Dolphin 2.6 Mistral 7B 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? 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