pchizhov / picky_bpe
BPE modification that implements removing of the intermediate tokens during tokenizer training.
☆25Updated 4 months ago
Alternatives and similar repositories for picky_bpe:
Users that are interested in picky_bpe are comparing it to the libraries listed below
- Code for SaGe subword tokenizer (EACL 2023)☆24Updated 4 months ago
- Trully flash implementation of DeBERTa disentangled attention mechanism.☆45Updated last week
- Repository containing the SPIN experiments on the DIBT 10k ranked prompts☆24Updated last year
- ☆44Updated 2 months ago
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆55Updated 7 months ago
- ☆33Updated 2 weeks ago
- QLoRA for Masked Language Modeling☆22Updated last year
- ReBase: Training Task Experts through Retrieval Based Distillation☆29Updated 2 months ago
- Code for PHATGOOSE introduced in "Learning to Route Among Specialized Experts for Zero-Shot Generalization"☆84Updated last year
- QAmeleon introduces synthetic multilingual QA data using PaLM, a 540B large language model. This dataset was generated by prompt tuning P…☆34Updated last year
- Code for the examples presented in the talk "Training a Llama in your backyard: fine-tuning very large models on consumer hardware" given…☆14Updated last year
- 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
- ☆54Updated 7 months ago
- ☆67Updated 8 months ago
- ☆48Updated 5 months ago
- ☆25Updated last year
- ☆23Updated 2 months ago
- Embedding Recycling for Language models☆38Updated last year
- ☆11Updated 4 months ago
- Code, results and other artifacts from the paper introducing the WildChat-50m dataset and the Re-Wild model family.☆29Updated 3 weeks ago
- Official Repository for "Hypencoder: Hypernetworks for Information Retrieval"☆23Updated last month
- 🕸 GlotCC Dataset and Pipline -- NeurIPS 2024☆18Updated 2 weeks ago
- Plug-and-play Search Interfaces with Pyserini and Hugging Face☆31Updated last year
- ☆13Updated this week
- Implementation of "SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models"☆27Updated 2 months ago
- Demonstration that finetuning RoPE model on larger sequences than the pre-trained model adapts the model context limit☆63Updated last year
- Reference implementation for Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model☆45Updated last year
- Training and evaluation code for the paper "Headless Language Models: Learning without Predicting with Contrastive Weight Tying" (https:/…☆26Updated last year
- Starbucks: Improved Training for 2D Matryoshka Embeddings☆19Updated 2 months ago
- Simple replication of [ColBERT-v1](https://arxiv.org/abs/2004.12832).☆80Updated last year