- Faster LLMs: Improving RWKV with Parallel Cumulative Sums
- Simple Fast Attention: Causal Implementation Experiments
- Negative Gearing and the CGT Discount: A Modern Portfolio Theory Analysis
- Retention LLMs: Analysing Algorithms and Alternative Implementations
- Digging into Doherty: Implications of Initialization
pipeline 1
- Deterministic Tensorflow Part 1: Model Training Dec 6, 2020