--- license: openrail datasets: - bigcode/the-stack language: - code programming_language: - Java - JavaScript - Python pipeline_tag: text-generation widget: - text: 'def print_hello_world():' example_title: Hello world group: Python model-index: - name: SantaCoder results: - task: type: text-generation dataset: type: openai_humaneval name: HumanEval (Python) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 verified: false - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL HumanEval (Java) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL HumanEval (JavaScript) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: openai_humaneval name: MBPP (Python) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 verified: false - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL MBPP (Java) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL MBPP (JavaScript) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: loubnabnl/humaneval_infilling name: HumanEval FIM (Python) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 verified: false - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL HumanEval FIM (Java) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: nuprl/MultiPL-E name: MultiPL HumanEval FIM (JavaScript) metrics: - name: pass@1 type: pass@1 value: 0.0 verified: false - name: pass@10 type: pass@10 value: 0.0 verified: false - name: pass@100 type: pass@100 value: 0.0 - task: type: text-generation dataset: type: code_x_glue_ct_code_to_text name: CodeXGLUE code-to-text (Python) metrics: - name: BLEU type: bleu value: 0.0 verified: false --- # SantaCoder ![banner](https://huggingface.co./datasets/bigcode/admin/resolve/main/banner.png) # Table of Contents 1. [Model Summary](#model-summary) 2. [Use](#use) 3. [Limitations](#limitations) 4. [Training](#training) 5. [Citation](#citation) # Model Summary The SantaCoder models are a series of 1B parameter models trained on Python, Java, and JavaScript. They were trained on datasets with different filter parameters and with architecture and objective variations. The main model uses multi-query attention, was trained using near-deduplication and commnent-to-code ratio as filtering criteria and using the Fill-in-the-Middle objective. - **Repository:** [bigcode/Megatron-LM](https://github.com/bigcode-project/Megatron-LM) - **Project Website:** [bigcode-project.org]www.bigcode-project.org) - **Paper:** [Coming soon]() - **Point of Contact:** [contact@bigcode-project.org](mailto:contact@bigcode-project.org) - **Languages:** Python, Java, and JavaScript |Model|Architecture|Objective|Filtering| |:-|:-|:-|:-| |`mha`|MHA|AR + FIM| Base | |`no-fim`| MQA | AR| Base | |`fim`| MQA | AR + FIM | Base | |`stars`| MQA | AR + FIM | GitHub stars | |`fertility`| MQA | AR + FIM | Tokenizer fertility | |`comments`| MQA | AR + FIM | Comment-to-code ratio | |`dedup-alt`| MQA | AR + FIM | Stronger near-deduplication | |`dedup-alt-comments`| MQA | AR + FIM | Stronger near-deduplication and comment-to-code ratio | The `dedup-alt-comments` model is the best performing model and was trained twice as long as the others. This checkpoint is the default model and available on the `main` branch. All other checkpoints are on separate branches with according names. # Use ## Intended use **Feel free to share your generations in the Community tab!** ## How to use ### Generation ```python # pip install -q transformers from transformers import AutoModelForCausalLM, AutoTokenizer checkpoint = "bigcode/santacoder" 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() inputs = tokenizer.encode("def print_hello_world():", return_tensors="pt").to(device) outputs = model.generate(inputs) print(tokenizer.decode(outputs[0])) ``` ### Fill-in-the-middle Fill-in-the-mid uses special tokens to identify the prefix/middle/suffic part of the input and output: ```python input_text = "def print_hello_world():\n \n print('Hello world!')" inputs = tokenizer.encode(input_text, return_tensors="pt").to(device) outputs = model.generate(inputs) print(tokenizer.decode(outputs[0])) ``` ### Load other checkpoints We upload the checkpoint of each experiment to a seperate branch as well as the intermediate checkpoints as commits on the branches. You can load them with the `revision` flag: ```python model = AutoModelForCausalLM.from_pretrained( "bigcode/santacoder", revision="no-fim", # name of branch or commit hash trust_remote_code=True ) ``` ### Attribution The pretraining dataset of the model was filtered for permissive licenses only. Nevertheless, the model can generate source code verbatim from the dataset which requires attribution. We provide a [search index](TODO) 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 in Python, Java, and JavaScript. 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. # Training ## Model - **Architecture:** GPT-2 model with multi-query attention and Fill-in-the-Middle objective - **Pretraining steps:** 600K - **Pretraining tokens:** 236 billion - **Precision:** float16 ## Hardware - **GPUs:** 96 Tesla V100 - **Training time:** 6.2 days - **Total FLOPS:** 2.1 x 10e21 ## Software - **Orchestration:** [Megatron-LM](https://github.com/bigcode-project/Megatron-LM) - **Neural networks:** [PyTorch](https://github.com/pytorch/pytorch) - **FP16 if applicable:** [apex](https://github.com/NVIDIA/apex) # Citation **TODO**