North Coast Synthesis Ltd.

Tag: LLaMA

RLHF vs. Diversity

2026-07-27 Video LLaMA alignment fine-tuning text Turning a base model into a chatbot or agent usually involves extensive Reinforcement Learning with Human Feedback (RLHF). Does that impair the model's ability to provide diverse output and avoid the Mad Libs phenomenon? Access: $$$ Pro

Automatic prompt injection

2026-07-20 Video LLaMA applications prompting text tokenization training utility Discrete optimization can be used to create universal prompt injections that subvert multiple different task prompts. Access: $ Basic

Truthiness-focused search

2026-02-09 Video LLaMA evaluation hallucination sampling text truthfulness It appears that the earlier, shallower layers of a transformer-type language model learn syntax, and later, deeper layers learn factual information. So can we boost factual accuracy by boosting the effect of deeper layers? I take the view that that's analogous to dosing the model with a mind-altering drug. Access: $$$ Pro

Quis custodiet reward models

2025-09-29 Video alignment training text LLaMA Gemma Large language models are "aligned" using smaller, specially trained reward models. These are often secret, and poorly studied even if public. This paper opens the door to exploring reward models by asking them about their values. Access: Free account

LLaMA introduction

2025-09-22 Video model-intro text LLaMA Facebook's entry into the LLM game: the first "open" version of LLaMA from 2023. This is a fairly conventional Transformer-type architecture, influential on the field because it created pressure for everybody to release weights of their announced models. Access: $$$ Pro

Better (than) tokenization with BLTs

2025-08-01 Video theory text LLaMA tokenization Using "patches" of input bytes, instead of a fixed token list, allows better scalability and improves performance on some tasks that are hard for token-based LLMs. Access: $ Basic