Llama2 70B Oasst Sft V10 by OpenAssistant

 ยป  All LLMs  ยป  OpenAssistant  ยป  Llama2 70B Oasst Sft V10   URL Share it on

  Autotrain compatible Dataset:argilla/databricks-dol...   Dataset:openassistant/oasst1 Dataset:rombodawg/losslessmega...   Dataset:shahules786/orca-best   En   Endpoints compatible   Llama   Pytorch   Region:us   Sft   Sharded

Llama2 70B Oasst Sft V10 Benchmarks

Llama2 70B Oasst Sft V10 (OpenAssistant/llama2-70b-oasst-sft-v10)

Llama2 70B Oasst Sft V10 Parameters and Internals

Model Type 
Causal decoder-only transformer language model
Use Cases 
Areas:
Research, Commercial applications
Applications:
Text-generation
Primary Use Cases:
Chat assistant applications
Limitations:
Model outputs may be unpredictable, inaccurate, biased, or objectionable.
Considerations:
Perform application-specific safety testing before deployment.
Additional Notes 
Embeddings padded to multiple of 128 for sharded inference compatibility.
Supported Languages 
en (full), de (limited), es (limited), fr (limited), it (limited), pt (limited), pl (limited), nl (limited), ro (limited), cs (limited), sv (limited)
Training Details 
Data Sources:
rombodawg/LosslessMegaCodeTrainingV2_1m_Evol_Uncensored, OpenAssistant/oasst1, shahules786/orca-best, argilla/databricks-dolly-15k-curated-multilingual
Methodology:
Fine-tuned in two stages: first on synthetic instructions and coding tasks, then on top human demonstrations.
Context Length:
4096
Hardware Used:
EPFL's Machine Learning and Optimization Laboratory, Natural Language Processing Lab
Model Architecture:
Causal decoder-only transformer architecture
Responsible Ai Considerations 
Fairness:
Testing mainly in English, outputs may be unpredictable in other scenarios.
Transparency:
Documented training processes and datasets.
Accountability:
Open-Assistant development team is accountable for model outputs.
Mitigation Strategies:
Developers should perform safety testing and tuning specific to their applications.
Input Output 
Input Format:
Prompt dialogue template with OpenAI's chatml format.
Accepted Modalities:
text
Output Format:
Assistant responses
Performance Tips:
Use the official Llama2 system message for improved inference.
LLM NameLlama2 70B Oasst Sft V10
Repository ๐Ÿค—https://huggingface.co/OpenAssistant/llama2-70b-oasst-sft-v10 
Model Size70b
Required VRAM138 GB
Updated2024-12-23
MaintainerOpenAssistant
Model Typellama
Model Files  9.8 GB: 1-of-15   9.8 GB: 2-of-15   10.0 GB: 3-of-15   9.8 GB: 4-of-15   9.8 GB: 5-of-15   9.8 GB: 6-of-15   10.0 GB: 7-of-15   9.8 GB: 8-of-15   9.8 GB: 9-of-15   9.8 GB: 10-of-15   10.0 GB: 11-of-15   9.8 GB: 12-of-15   9.8 GB: 13-of-15   9.5 GB: 14-of-15   0.5 GB: 15-of-15
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.32.1
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32007
Torch Data Typebfloat16

Quantized Models of the Llama2 70B Oasst Sft V10

Model
Likes
Downloads
VRAM
Llama2 70B OASST SFT V10 GPTQ4409235 GB
Llama2 70B OASST SFT V10 GGUF1027829 GB
Llama2 70B OASST SFT V10 AWQ24036 GB
Llama2 70B OASST SFT V10 GGML41629 GB

Best Alternatives to Llama2 70B Oasst Sft V10

Best Alternatives
Context / RAM
Downloads
Likes
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB31875
... 3 70B Instruct Gradient 1048K1024K / 141.9 GB276121
Llama3 Function Calling 1048K1024K / 141.9 GB31
...a 3 70B Instruct Gradient 524K512K / 141.9 GB5423
...a 3 70B Instruct Gradient 262K256K / 141.9 GB19955
...ama 3 70B Arimas Story RP V2.0256K / 141.1 GB523
...ama 3 70B Arimas Story RP V1.6256K / 141.2 GB160
...ama 3 70B Arimas Story RP V1.5256K / 141.2 GB312
Yi 70B 200K RPMerge Franken195K / 142.4 GB181
...a 3.1 Nemotron 70B Instruct HF128K / 141.9 GB1568661929
Note: green Score (e.g. "73.2") means that the model is better than OpenAssistant/llama2-70b-oasst-sft-v10.

Rank the Llama2 70B Oasst Sft V10 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? 40126 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217