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

Code 290K 6.7B Instruct by ajibawa-2023

 »  All LLMs  »  ajibawa-2023  »  Code 290K 6.7B Instruct   URL Share it on

  87   
  Autotrain compatible   Code   Codegen   Conversational Dataset:ajibawa-2023/code-290k...   En   Endpoints compatible   Instruct   Llama   Model-index   Region:us   Safetensors   Sharded   Tensorflow

Code 290K 6.7B Instruct Benchmarks

ARC: 34.9 vs 96.7 (so35)-63.9%
HellaSwag: 51.99 vs 95.3 (gpt4)-45.4%
MMLU: 34.89 vs 88.3 (so35)-60.5%
TruthfulQA: 41.95 vs 59 (gpt4)-28.9%
WinoGrande: 52.64 vs 87.5 (gpt4)-39.8%
GSM8K: 3.49 vs 96.4 (so35)-96.4%
LLME Score: 0.22829
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Code 290K 6.7B Instruct Parameters and Internals

Model Type 
text generation, code generation
Use Cases 
Areas:
Research, Commercial applications
Primary Use Cases:
Code generation, Providing detailed code explanations
Additional Notes 
Quantized version available thanks to Bartowski
Supported Languages 
Python (yes), Java (yes), JavaScript (yes), GO (yes), C++ (yes), Rust (yes), Ruby (yes), Sql (yes), MySql (yes), R (yes), Julia (yes), Haskell (yes)
Training Details 
Data Sources:
Code-290k-ShareGPT dataset
Data Volume:
290,000 sets of codes
Methodology:
Fine-tuning
Training Time:
85 hours (3 epochs)
Hardware Used:
4 x A100 80GB GPUs
Input Output 
Input Format:
Alpaca format
Accepted Modalities:
text
Output Format:
text with code and explanations
Performance Tips:
Modify prompt as needed for best results.
LLM NameCode 290K 6.7B Instruct
Repository 🤗https://huggingface.co/ajibawa-2023/Code-290k-6.7B-Instruct 
Model Size6.7b
Required VRAM13.5 GB
Updated2024-11-22
Maintainerajibawa-2023
Model Typellama
Instruction-BasedYes
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   3.6 GB: 3-of-3   0.0 GB
Supported Languagesen
Generates CodeYes
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length16384
Model Max Length16384
Transformers Version4.37.2
Tokenizer ClassLlamaTokenizer
Padding Token<｜end▁of▁sentence｜>
Vocabulary Size32256
Torch Data Typebfloat16
Code 290K 6.7B Instruct (ajibawa-2023/Code-290k-6.7B-Instruct)

Best Alternatives to Code 290K 6.7B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
...s Coder6.7b Reflct Adamw Iter116K / 13.5 GB3200
Speechless Coder Ds 6.7B16K / 13.5 GB12685
Datascience Coder 6.7B16K / 13.5 GB12463
Hpc Coder V2.6.7B16K / 27.1 GB2716
Merge Code Gen16K / 13.6 GB140
... 6.7B Instruct Trt Int4 G64 Hf16K / 4.2 GB60
...6.7B Instruct Trt Int8 G128 Hf16K / 7.2 GB60
NaturalSQL 6.7B V016K / 13.5 GB6346
SQL Sorcerer 6.7B V1.116K / 13.5 GB80
DeepYi Second16K / 12.6 GB170
Note: green Score (e.g. "73.2") means that the model is better than ajibawa-2023/Code-290k-6.7B-Instruct.

Rank the Code 290K 6.7B 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? 38199 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 v20241110
  Support LLM Explorer