evintunador / next-concept-predictorLinks
a WIP architecture designed to allow transformers to think in a manner without tokens
☆19Updated last year
Alternatives and similar repositories for next-concept-predictor
Users that are interested in next-concept-predictor are comparing it to the libraries listed below
Sorting:
- Training an LLM to use a calculator with multi-turn reinforcement learning, achieving a **62% absolute increase in evaluation accuracy**.☆41Updated last month
- ☆38Updated 11 months ago
- Optimizing Causal LMs through GRPO with weighted reward functions and automated hyperparameter tuning using Optuna☆53Updated 4 months ago
- LLM reads a paper and produce a working prototype☆57Updated 2 months ago
- ☆47Updated last year
- ☆21Updated 3 weeks ago
- ☆20Updated 3 months ago
- Official homepage for "Self-Harmonized Chain of Thought" (NAACL 2025)☆91Updated 5 months ago
- Simple repository for training small reasoning models☆33Updated 4 months ago
- Simple GRPO scripts and configurations.☆58Updated 4 months ago
- ☆63Updated last month
- look how they massacred my boy☆63Updated 8 months ago
- LLM training in simple, raw C/CUDA☆14Updated 6 months ago
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆101Updated 3 months ago
- The original BabyAGI, updated with LiteLLM and no vector database reliance (csv instead)☆21Updated 8 months ago
- An intelligent code optimization system leveraging AI analysis, automated refactoring, and test generation. Built with DSPy and Gradio, i…☆20Updated 4 months ago
- Automated Capability Discovery via Foundation Model Self-Exploration☆52Updated 4 months ago
- ☆30Updated 7 months ago
- Testing paligemma2 finetuning on reasoning dataset☆18Updated 5 months ago
- Code and data for the paper "Why think step by step? Reasoning emerges from the locality of experience"☆60Updated 2 months ago
- Latent Large Language Models☆18Updated 10 months ago
- KMD is a collection of conversational exchanges between patients and doctors on various medical topics. It aims to capture the intricaci…☆24Updated last year
- An example implementation of RLHF (or, more accurately, RLAIF) built on MLX and HuggingFace.☆29Updated last year
- A fast, local, and secure approach for training LLMs for coding tasks using GRPO with WebAssembly and interpreter feedback.☆30Updated 2 months ago
- j1-micro (1.7B) & j1-nano (600M) are absurdly tiny but mighty reward models.☆82Updated 3 weeks ago
- ☆47Updated 4 months ago
- Lego for GRPO☆28Updated last month
- A framework for pitting LLMs against each other in an evolving library of games ⚔☆32Updated 2 months ago
- A reading list of relevant papers and projects on foundation model annotation☆27Updated 3 months ago
- Writing Blog Posts with Generative Feedback Loops!☆48Updated last year