pchizhov / picky_bpe
BPE modification that implements removing of the intermediate tokens during tokenizer training.
☆13Updated last week
Related projects: ⓘ
- Plug-and-play Search Interfaces with Pyserini and Hugging Face☆32Updated last year
- Embedding Recycling for Language models☆38Updated last year
- ☆18Updated this week
- QAmeleon introduces synthetic multilingual QA data using PaLM, a 540B large language model. This dataset was generated by prompt tuning P…☆33Updated last year
- Training and evaluation code for the paper "Headless Language Models: Learning without Predicting with Contrastive Weight Tying" (https:/…☆22Updated 5 months ago
- SWIM-IR is a Synthetic Wikipedia-based Multilingual Information Retrieval training set with 28 million query-passage pairs spanning 33 la…☆42Updated 10 months ago
- PyTorch implementation for MRL☆17Updated 6 months ago
- QLoRA for Masked Language Modeling☆20Updated last year
- Repository containing the SPIN experiments on the DIBT 10k ranked prompts☆22Updated 6 months ago
- Code for SaGe subword tokenizer (EACL 2023)☆21Updated this week
- ☆29Updated 3 weeks ago
- Library for fast text representation and classification.☆28Updated 8 months ago
- ReBase: Training Task Experts through Retrieval Based Distillation☆27Updated 2 months ago
- Index of URLs to pdf files all over the internet and scripts☆20Updated last year
- Using short models to classify long texts☆20Updated last year
- ☆12Updated last week
- ☆24Updated last year
- a pipeline for using api calls to agnostically convert unstructured data into structured training data☆26Updated last year
- Using open source LLMs to build synthetic datasets for direct preference optimization☆33Updated 6 months ago
- ☆29Updated 2 weeks ago
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆39Updated 2 weeks ago
- This is a new metric that can be used to evaluate faithfulness of text generated by LLMs. The work behind this repository can be found he…☆31Updated last year
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆62Updated last year
- Simple replication of [ColBERT-v1](https://arxiv.org/abs/2004.12832).☆73Updated 6 months ago
- Minimum Bayes Risk Decoding for Hugging Face Transformers☆51Updated 3 months ago
- Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model☆40Updated 8 months ago
- Improving Text Embedding of Language Models Using Contrastive Fine-tuning☆54Updated last month
- ☆68Updated last month
- ☆44Updated 2 months ago
- minimal pytorch implementation of bm25 (with sparse tensors)☆82Updated 6 months ago