Language Models are Changing AI: The Need for Holistic Evaluation. New benchmarking method, Holistic Evaluation of Language Models (HELM) has been developed at the Center for Research on Foundation Models to help provide transparency in LMs. @RishiBommasani
, @PercyLiang
, Tony Lee
GENERATIVE AI
-
HELM: Holistic Evaluation Method for Language Models
By
–
-
Teaching Algorithmic Reasoning to LLMs via In-context Learning
By
–
Teaching Algorithmic Reasoning via In-context Learning. A four-stage approach to teaching algorithmic reasoning to LLMs is identified and studied in this work by @oh_that_hat
, @Azade_Na
, @Hugo_Larochelle
, Aaron Courville, @BNeyshabur
, @HanieSedghi -
Intermediate Models Value Through Transfer Learning Fine-tuning
By
–
Where to start? Analyzing the potential value of intermediate models. By finetuning a model on a source dataset, one may have a better starting point when finetuning a target dataset. @LChoshen
, Elad Venezian, Shachar Don-Yehia, @NoamSlonim
, @YoavKatz -
LLMs Struggle Learning Long-Tail Knowledge From Training Data
By
–
Large Language Models Struggle to Learn Long-Tail Knowledge. @kandpal_nikhil
, @HaikangDeng
, Adam Roberts, @Eric_Wallace_
, @ColinRaffel examine the relationship between their pre-training datasets and their knowledge memorized. -
Hyperwrite AI Chatbot Alternative to ChatGPT Now Available
By
–
While OpenAI ChatGPT is under heavy load, you can use our version! The UI might not be as pretty (yet), but with no signup restrictions, you can use it NOW to help you with your writing and pretty much anything else. Try it here: https://
hyperwriteai.com/chatbot -
Configurable AI Models and Safety Mode Control Options
By
–
i think it’s very reasonable for “default models” to be inoffensive, but within some pretty bounds i think people should be able to configure the models to behave how they’d like, eg like turning off safe mode
-
Pivoting Strategy: Improving Results Through Better Prompt Architecture
By
–
so now I am pivoting and basically just going to try better prompt architecture within the application to get good results, I had some fun playing around with simple prompts like this already so I know it could work somewhat well
-
Fine-tuned Models Producing Poor Results: Data Quality Issues
By
–
but the results from the fine-tuned models were trash and the responses I was getting were really bad I am assuming it takes either significantly more refined prompt/completion data than what I gave it or just more data (both could be true)
-
Initial Fine-Tuning Approach: Breaking Books into Chunks and Generating Questions
By
–
initial approach: oh this seems not too bad, I read this doc https://
beta.openai.com/docs/guides/fi
ne-tuning
… and was like yeah I can just break up the authors' books into chunks and generate some simple questions for each chunk and then use that to fine-tune the model -
Philosophy Author Chatbot Project Using GPT-3 Fine-Tuning
By
–
philosophy author chatbot thing w gpt background: need to create something for my philosophy of AI class idea: fine-tune gpt3 models on text from authors we have read in class and create a chat application where users can talk to these models