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

Llama 3 8B Instruct 262K by gradientai

 »  All LLMs  »  gradientai  »  Llama 3 8B Instruct 262K   URL Share it on

  13182   
  Arxiv:2305.14233   Arxiv:2309.00071   Arxiv:2402.08268   Autotrain compatible   Conversational   En   Endpoints compatible   Instruct   Llama   Llama-3   Meta   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/gradientai/Llama-3-8B-Instruct-262k 

Llama 3 8B Instruct 262K Benchmarks

ARC: 53.16 vs 96.7 (so35)-45%
HellaSwag: 75.53 vs 95.3 (gpt4)-20.7%
MMLU: 64.27 vs 88.3 (so35)-27.2%
TruthfulQA: 48.37 vs 59 (gpt4)-18%
WinoGrande: 73.01 vs 87.5 (gpt4)-16.6%
GSM8K: 47.23 vs 96.4 (so35)-51%
LLME Score: 0.25909
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
🌟 Advertise your project 🚀

Llama 3 8B Instruct 262K Parameters and Internals

Model Type 
text generation, dialogue, instruction tuned
Use Cases 
Areas:
Commercial use, Research use
Applications:
Dialog and assistant-like applications
Primary Use Cases:
Assistant-like chat and instruction tasks
Limitations:
Inapplicable in languages other than English without additional tuning., Requires appropriate safety tuning for specialized applications.
Considerations:
Developers should employ additional safeguarding measures and consider context when deploying applications.
Additional Notes 
Quantized versions available for resource-efficient deployments.
Supported Languages 
en (Primary support for dialogue and instructional tasks in English.)
Training Details 
Data Sources:
publicly available online data, SlimPajama-627B, UltraChat
Data Volume:
15 trillion tokens (pretraining)
Methodology:
Supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF). NTK-aware interpolation technique for context length adjustment.
Context Length:
262144
Training Time:
Total of 7.7M GPU hours for pretraining across multiple models.
Hardware Used:
Crusoe Energy high performance L40S cluster
Model Architecture:
Optimized transformer using RoPE theta for extended contexts.
Safety Evaluation 
Methodologies:
red-teaming, adversarial evaluations
Findings:
Improved refusal rate for false prompts in comparison to Llama 2., Limited residual risks assessed through community tools.
Risk Categories:
Child safety, Cybersecurity vulnerabilities, CBRNE threats
Ethical Considerations:
Guided by a Responsible Use Guide and supporting tools like Meta Llama Guard 2.
Responsible Ai Considerations 
Fairness:
Focused on openness, inclusivity, and helpfulness while respecting diverse values and perspectives.
Transparency:
Incorporates feedback from the community to continually assess safety and alignment.
Accountability:
Developers undertaking use are held to the standards under the Acceptable Use Policy.
Mitigation Strategies:
Use community safeguards like Llama Guard to supplement model-level safety.
Input Output 
Input Format:
Text input following a conversational template.
Accepted Modalities:
text
Output Format:
Generated text and code responses.
Performance Tips:
Leverage context window optimizations for handling extensive lengths.
Release Notes 
Version:
5/3 Update
Date:
2024-05-03
Notes:
Finalized assistant-like chat optimizations and context length extension techniques.
LLM NameLlama 3 8B Instruct 262K
Repository 🤗https://huggingface.co/gradientai/Llama-3-8B-Instruct-262k 
Model Size8b
Required VRAM16.1 GB
Updated2025-06-11
Maintainergradientai
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length262144
Model Max Length262144
Transformers Version4.41.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16
Llama 3 8B Instruct 262K (gradientai/Llama-3-8B-Instruct-262k)

Quantized Models of the Llama 3 8B Instruct 262K

Model
Likes
Downloads
VRAM
...Instruct 262K HQQ 2bit Smashed2334 GB
Llama 3 8B Instruct 262K GGUF2413 GB

Best Alternatives to Llama 3 8B Instruct 262K

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB3715108
UltraLong Thinking4192K / 16.1 GB8292
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB92315
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
Zero Llama 3.1 8B Beta61048K / 16.1 GB14161
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB207045
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
....1 1million Ctx Dark Planet 8B1048K / 32.3 GB952
...dger Nu Llama 3.1 8B UltraLong1048K / 16.2 GB133
Note: green Score (e.g. "73.2") means that the model is better than gradientai/Llama-3-8B-Instruct-262k.

Rank the Llama 3 8B Instruct 262K 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? 48079 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 v20241124
  Support LLM Explorer