Hugging Face's logo

bigcode
/
starcoder2-15b

Text Generation
Transformers
Safetensors
starcoder2
code
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
26
starcoder2-15b / README.md
lvwerra's picture
lvwerra HF staff
Update README.md
76e8fbd verified
preview code
|
raw
history blame
3.69 kB
metadata
pipeline_tag: text-generation
inference: true
widget:
  - text: 'def print_hello_world():'
    example_title: Hello world
    group: Python
datasets:
  - bigcode/the-stack-v2-train
license: bigcode-openrail-m
library_name: transformers
tags:
  - code

StarCoder

TODO banner

Table of Contents

  1. Model Summary
  2. Use
  3. Limitations
  4. Training
  5. License
  6. Citation

Model Summary

The StarCoderBase models are 15.5B parameter models trained on 600+ programming languages from The Stack v2, with opt-out requests excluded. The model uses Grouped Query Attention, a context window of 16,384 tokens with a sliding window attention of 4,096 tokens, and was trained using the Fill-in-the-Middle objective on 4+ trillion tokens.

Use

Intended use

The model was trained on GitHub code as well as additional selected data sources such as Arxiv and Wikipedia. As such it is not an instruction model and commands like "Write a function that computes the square root." do not work well.

Generation 

# pip install -q transformers # TODO: from main
from transformers import AutoModelForCausalLM, AutoTokenizer

checkpoint = "bigcode/starcoderbase"
device = "cuda" # for GPU usage or "cpu" for CPU usage

tokenizer = AutoTokenizer.from_pretrained(checkpoint)
model = AutoModelForCausalLM.from_pretrained(checkpoint, trust_remote_code=True).to(device)

inputs = tokenizer.encode("def print_hello_world():", return_tensors="pt").to(device)
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))

Attribution & Other Requirements

The pretraining dataset of the model was filtered for permissive licenses and code with no license only. Nevertheless, the model can generate source code verbatim from the dataset. The code's license might require attribution and/or other specific requirements that must be respected. We provide a search index that let's you search through the pretraining data to identify where generated code came from and apply the proper attribution to your code.

Limitations

The model has been trained on source code from 600+ programming languages. The predominant language in source is English although other languages are also present. As such the model is capable to generate code snippets provided some context but the generated code is not guaranteed to work as intended. It can be inefficient, contain bugs or exploits. See the paper for an in-depth discussion of the model limitations.

Training

Model

  • Architecture: Transformer decoder with grouped-query and sliding window attention and Fill-in-the-Middle objective
  • Pretraining steps: TODO
  • Pretraining tokens: 4+ trillion
  • Precision: bfloat16

Hardware

  • GPUs: 1024 A100
  • Training time: TODO

Software

License

The model is licensed under the BigCode OpenRAIL-M v1 license agreement. You can find the full agreement here.

Citation

TODO