Re:Can GPT-4 and GPT-3.5 play Wordle?
by og_kalu on Monday March 20, 2023 @10:51PM
(#35241983)
Nobody actually really knows what LLMs or ML models in general learn.
This is a high level overview of basic training of a model.
You give it a task and a structure (the architecture) to learn that task. The model tries to perform the task. It fails horribly. However, it gains a little bit of of something from the attempt. That something is used to adjust the weights/neurons/parameters etc. What some misunderstand a bit about weights...they're not copies of the training data. What i mean is, the model is going to be the same sie whether it's trained on 2 images or 200 million or 2 billion. You can think of parameters as configuration settings of sorts. after every botched attempt, the settings get tuned. on and on and on until training is stopped for whatever reason.
What is the model learning ? What is happening to the weights exactly? What is being shifted and why ? What are the weights actually doing ?
This is a high level overview of basic training of a model.
You give it a task and a structure (the architecture) to learn that task. The model tries to perform the task. It fails horribly. However, it gains a little bit of of something from the attempt. That something is used to adjust the weights/neurons/parameters etc. What some misunderstand a bit about weights...they're not copies of the training data. What i mean is, the model is going to be the same sie whether it's trained on 2 images or 200 million or 2 billion. You can think of parameters as configuration settings of sorts. after every botched attempt, the settings get tuned. on and on and on until training is stopped for whatever reason.
What is the model learning ? What is happening to the weights exactly? What is being shifted and why ? What are the weights actually doing ?
Excellent questions. Too bad nobody knows.