Gpt Neo 125M Code Clippy Dedup 2048

🚀 gpt-neo-125M-code-clippy-dedup-2048

This is a text generation model developed by the Flax Community and shared by Hugging Face, suitable for text generation tasks.

🚀 Quick Start

Use the code below to get started with the model.

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("flax-community/gpt-neo-125M-code-clippy-dedup-2048")

model = AutoModelForCausalLM.from_pretrained("flax-community/gpt-neo-125M-code-clippy-dedup-2048")

✨ Features

This model can be used for the task of Text Generation.

📚 Documentation

Model Details

• Developed by: Flax Community
• Shared by [Optional]: Hugging Face
• Model type: Text Generation
• Related Models:
  • Parent Model: GPT - Neo
• Resources for more information:
  • GitHub Repo

Uses

Direct Use

This model can be used for the task of Text Generation.

Out-of-Scope Use

The model should not be used to intentionally create hostile or alienating environments for people.

Bias, Risks, and Limitations

Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.

Recommendations

The model creators note in the GitHub Repo:

ISSUE : Wrong Filenames in the Dataset We recently came to know about a bug which happened during the scraping of the dataset. We found out that the file names are obsolete/misleading. [Refer this issue] We thank Naman for pointing out the issue. This might have two implications: - Since the filtering for the training dataset is done using the file extension, we might have had wrong datapoints in the dataset while training and we might have missed a lot of right datapoints that belong to the languages of choice.

Training Details

Training Data

The model creators note in the GitHub Repo:

For fine-tuning GPTNeo - 125M on CodeClippy dataset we used AdamW optimizer (beta1 = 0.9, beta2 = 0.95) with GPT3 - like learning rate schedule (4k warmup steps from 0 to 5e - 5 followed by 50k cosine decay steps to 5e - 6), weight decay 0.1 and batch size 1024, sequence length 2048.

Training Procedure

Speeds, Sizes, Times

The model creators note in the GitHub Repo:

For fine-tuning GPTNeo - 125M on CodeClippy dataset we used AdamW optimizer (beta1 = 0.9, beta2 = 0.95) with GPT3 - like learning rate schedule (4k warmup steps from 0 to 5e - 5 followed by 50k cosine decay steps to 5e - 6), weight decay 0.1 and batch size 1024, sequence length 2048. The choice of relatively large batch size and low LR with long warmup are made to avoid agressive updates and preserve the knowledge contained in pretrained GPTNeo weights.

Evaluation

Testing Data, Factors & Metrics

Testing Data

The model creators note in the GitHub Repo:

The models are also evaluated on the APPS and HumanEval datasets.

Results

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Technical Specifications

Model Architecture and Objective

GPTNeoForCausalLM

Citation

Model Card Authors: Flax Community in collaboration with Ezi Ozoani and the Hugging Face team