It's not just about compute though, the tokens have some meaning to them. It doesn't work to just get the model to generate arbitrary tokens to increase compute. I would be skeptical of this working even if you trained the model on such examples.
LLMS
-
LangChain Question Answering Instructions Tutorial
By
–
try these instructions? https://
python.langchain.com/en/latest/use_
cases/question_answering.html
… -
Microsoft’s Automatic Prompt Optimization Boosts LLM Performance
By
–
Microsoft’s Automatic Prompt Optimization Improves Prompts to Boost LLM Performance https://
syncedreview.com/2023/05/09/mic
rosofts-automatic-prompt-optimization-improves-prompts-to-boost-llm-performance/
… -
Study on Chain-of-Thought Faithfulness and Model Bias
By
–
Really cool paper studying the faithfulness of chain-of-thought (CoT): http://
arxiv.org/abs/2305.04388 The paper uses a biased prompt to try to mislead the model. For example, all few-shot exemplars could be answer (A), or they could add a suffix such as "I think the answer is but -
When Not to Trust Language Models: Memory Limitations
By
–
11/ When Not to Trust Language Models: Investigating Effectiveness and Limitations of Parametric and Non-Parametric Memories Authors: @alextmallen
, @AkariAsai
, @hllo_wrld
, Rajarshi Das, @HannaHajishirzi
, @DanielKhashabi
, -
Wendy’s Deploys Custom Language Model for Burger Ordering
By
–
Wendy’s customized language model includes unique terms, phrases and acronyms customers have come to use when ordering its burgers, fries and other items—such as “JBC” for junior bacon cheeseburger…
-
GPT-4 Interpretability Research on GPT-2 Models
By
–
GPT-4 doing some interpretability work on GPT-2:
-
Karpathy defines new paradigm for improving large language models
By
–
leave it to @karpathy to help define the entire paradigm for improving LLMs 🙂
-
AI Enabling Conversations with History’s Greatest Meditation Teachers
By
–
the best part is AI will help us in these endeavors
— Alex Albert (@alexalbert__) 9 mai 2023
“imagine being able to talk with the best meditation teacher in history” – @ilyasut pic.twitter.com/hZCZKTlHBQthe best part is AI will help us in these endeavors “imagine being able to talk with the best meditation teacher in history” – @ilyasut
-
ChatGPT and WolframAlpha: Truth, Knowledge, and Intelligence
By
–
Here's my conversation with @stephen_wolfram
, his 4th time on the podcast. This was a deep and fascinating exploration of ChatGPT and WolframAlpha as it relates to the nature of truth, knowledge, intelligence, entropy, complexity, and reality itself. https://
youtube.com/watch?v=PdE-wa
Sx-d8
…