4 ms·

Maybe it really does have seasonal depression. The current date is part of the system prompt as far as I can tell. At least on ChatGPT you can ask for current

by kaoD 3y ago

Maybe it really does have seasonal depression. The current date is part of the system prompt as far as I can tell. At least on ChatGPT you can ask for current date, not sure about API (EDIT: indeed, this is explained further in the tweet that OP links to[0], go there since that's where the actual info is). Maybe the model learnt that humans in general tend to give shorter/lazier responses around this date (being trained on texts with dates in metadata, think forum posts, blogs, tweets, etc.) so it could very well be imitating trends in seasonal human behavior. [0] https://nitter.net/RobLynch99/status/1734278713762549970 https://nitter.net/RobLynch99/status/1734278713762549970

johnea 3y ago
Or, maybe the human species has totally lost the concept of physical reality...
AndrewKemendo 3y ago
In fact, it should be expected. The underlying data is embedded with artifacts of behaviors/affects that reflect the underlying state of the person who wrote them. Stanford famously studied this for reddit (notably a huge part of the commoncrawl dataset) and reported that it had non-trivial percentages of anti-social sentiments embedded in the text [1] You can't filter out the biases embedded in the data because it is a functional artifact of what the people creating the data were communicating. Best you can do is put guardrails and censor it, but that's a neverending game you can't win. Junk in Junk out still applies to LLMs [1]https://hci.stanford.edu/publications/2022/Park_ContentModAudit_CSCW22.pdf https://hci.stanford.edu/publications/2022/Park_ContentModAu...
- madsbuch 3y ago
  Not all content comes associated with a date. I reckon enough does, that a signal is embedded in the model.
  - AndrewKemendo 3y ago
    I’ll bet it’s really friendly around Christmas I wonder how we would test it
    - vineyardmike 3y ago
      Well the original post uses a for loop with a prompt that manually sets the date to various dates, then measures the length. So it seems very easy to set a holiday date. Testing “friendliness” or “holiday cheer” could be done via some sortof proxy… you could prompt it to reply to generic pleasantries while role playing “as someone busy who is late” and see if the word choice changes. Or maybe ask it to be a judge for a crime of desperation (stealing toys for orphans?) and determine sentencing durations as a proxy for sympathy? I suspect that is too far from training data though.
dang 3y ago
(This comment was originally a reply to https://news.ycombinator.com/item?id=38604519 https://news.ycombinator.com/item?id=38604519, but we merged that thread hither.)

by kaoD 3y ago

johnea 3y ago

Or, maybe the human species has totally lost the concept of physical reality...

AndrewKemendo 3y ago

In fact, it should be expected. The underlying data is embedded with artifacts of behaviors/affects that reflect the underlying state of the person who wrote them. Stanford famously studied this for reddit (notably a huge part of the commoncrawl dataset) and reported that it had non-trivial percentages of anti-social sentiments embedded in the text [1] You can't filter out the biases embedded in the data because it is a functional artifact of what the people creating the data were communicating. Best you can do is put guardrails and censor it, but that's a neverending game you can't win. Junk in Junk out still applies to LLMs [1]https://hci.stanford.edu/publications/2022/Park_ContentModAudit_CSCW22.pdf https://hci.stanford.edu/publications/2022/Park_ContentModAu...

madsbuch 3y ago
Not all content comes associated with a date. I reckon enough does, that a signal is embedded in the model.
- AndrewKemendo 3y ago
  I’ll bet it’s really friendly around Christmas I wonder how we would test it
  - vineyardmike 3y ago
    Well the original post uses a for loop with a prompt that manually sets the date to various dates, then measures the length. So it seems very easy to set a holiday date. Testing “friendliness” or “holiday cheer” could be done via some sortof proxy… you could prompt it to reply to generic pleasantries while role playing “as someone busy who is late” and see if the word choice changes. Or maybe ask it to be a judge for a crime of desperation (stealing toys for orphans?) and determine sentencing durations as a proxy for sympathy? I suspect that is too far from training data though.

dang 3y ago

(This comment was originally a reply to https://news.ycombinator.com/item?id=38604519 https://news.ycombinator.com/item?id=38604519, but we merged that thread hither.)