Norod / TrainGPT2-127M-FromScratch
A trio of Google-Colab notebooks (ipynb) for training a GPT-2 (127M) model from scratch (useful for other / non-English languages) using gpt-2-simple
☆15Updated 4 years ago
Related projects ⓘ
Alternatives and complementary repositories for TrainGPT2-127M-FromScratch
- ☆48Updated last year
- ☆42Updated last year
- Hidden Engrams: Long Term Memory for Transformer Model Inference☆34Updated 3 years ago
- Low-Rank Adaptation of Large Language Models clean implementation☆9Updated last year
- Text-writing denoising diffusion (and much more)☆30Updated last year
- Load any clip model with a standardized interface☆21Updated 6 months ago
- Experimental sampler to make LLMs more creative☆30Updated last year
- A client library for LAION's effort to filter CommonCrawl with CLIP, building a large scale image-text dataset.☆31Updated last year
- ☆32Updated last year
- One stop shop for all things carp☆59Updated 2 years ago
- Conversational Language model toolkit for training against human preferences.☆41Updated 7 months ago
- ☆24Updated last year
- Using short models to classify long texts☆20Updated last year
- A public implementation of the ReLoRA pretraining method, built on Lightning-AI's Pytorch Lightning suite.☆33Updated 8 months ago
- SCREWS: A Modular Framework for Reasoning with Revisions☆26Updated last year
- ☆31Updated 10 months ago
- GPT-jax based on the official huggingface library☆13Updated 3 years ago
- Fine-tune and quantize Llama-2-like models to generate Python code using QLoRA, Axolot,..☆64Updated 9 months ago
- ☆14Updated last year
- jupyter/colab implementation of stable-diffusion using k_lms sampler, cpu draw manual seeding, and quantize.py fix☆38Updated 2 years ago
- GreenLIT: Using GPT-J with Multi-Task Learning to Create New Screenplays☆17Updated last year
- Consists of the largest (10K) human annotated code-switched semantic parsing dataset & 170K generated utterance using the CST5 augmentati…☆33Updated last year
- Make GPT-2 complete your text in Colab. Use fancy JS, instead of default ugly colab interface.☆19Updated 3 years ago
- A repository re-creating the PromptBreeder Evolutionary Algorithm from the DeepMind Paper in Python using LMQL as the backend.☆27Updated last year
- ☆27Updated last year
- Image restoration with neural networks but without learning.☆46Updated 2 years ago
- 🐜🔧 A minimalistic tool to fine-tune your LLMs☆17Updated last year
- A ready-to-deploy container for implementing an easy to use REST API to access Language Models.☆64Updated last year
- Create soft prompts for fairseq 13B dense, GPT-J-6B and GPT-Neo-2.7B for free in a Google Colab TPU instance☆27Updated last year