<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p><br>
</p>
<div class="moz-cite-prefix">On 4/7/23 10:48 AM, Stephen Guerin
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAOmOqnJ7K4gNsmWszmmh0U_SGv9Wswu6mfNcCdeBGbrCp546Mg@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="auto">
<div dir="auto">As I wrote the prompt, I was wondering how it
might deal with the ambiquity of color for these objects.</div>
<div dir="auto"><br>
</div>
Prompt: <span style="color:rgb(236,236,241);font-family:söhne,ui-sans-serif,"system-ui",-apple-system,"segoe ui",roboto,ubuntu,cantarell,"noto sans",sans-serif,"helvetica neue",arial,"apple color emoji","segoe ui emoji","segoe ui symbol","noto color emoji";font-size:16px;white-space:pre-wrap;background-color:rgb(52,53,65)">What might a fire engine, an apple and a rose have in common</span>
<div dir="auto"><span style="color:rgb(236,236,241);font-family:söhne,ui-sans-serif,"system-ui",-apple-system,"segoe ui",roboto,ubuntu,cantarell,"noto sans",sans-serif,"helvetica neue",arial,"apple color emoji","segoe ui emoji","segoe ui symbol","noto color emoji";font-size:16px;white-space:pre-wrap;background-color:rgb(52,53,65)">
</span></div>
<div dir="auto"><span style="color:rgb(236,236,241);font-family:söhne,ui-sans-serif,"system-ui",-apple-system,"segoe ui",roboto,ubuntu,cantarell,"noto sans",sans-serif,"helvetica neue",arial,"apple color emoji","segoe ui emoji","segoe ui symbol","noto color emoji";font-size:16px;white-space:pre-wrap;background-color:rgb(52,53,65)">ChatGPT4:</span></div>
<div dir="auto"><span style="color:rgb(209,213,219);font-family:söhne,ui-sans-serif,"system-ui",-apple-system,"segoe ui",roboto,ubuntu,cantarell,"noto sans",sans-serif,"helvetica neue",arial,"apple color emoji","segoe ui emoji","segoe ui symbol","noto color emoji";font-size:16px;white-space:pre-wrap;background-color:rgb(68,70,84)">A fire engine, an apple, and a rose might not appear to have much in common at first glance, as they belong to different categories\u2014vehicles, fruits, and flowers, respectively. However, one possible commonality among them is their color: red. Fire engines are typically red, apples can be red (among other colors), and roses can also be red. The color red is often associated with strong emotions such as love, passion, and urgency, which may be another connection among these seemingly unrelated object</span><br>
</div>
</div>
</blockquote>
<p>After reading the article and dipping into the paper, I asked
chatGPT4 about the differences of color perception, naming and
organization across cultures, the answer was pretty mundane and
mealy-mouthed IMO. I was hoping for a more subtle or extravagant
treatment. I will be trying to develop my "prompt engineer"
skills more based on this topic...</p>
<p><br>
</p>
<p><br>
</p>
<blockquote type="cite"
cite="mid:CAOmOqnJ7K4gNsmWszmmh0U_SGv9Wswu6mfNcCdeBGbrCp546Mg@mail.gmail.com"><br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Fri, Apr 7, 2023, 10:15 AM
Steve Smith <<a href="mailto:sasmyth@swcp.com"
moz-do-not-send="true" class="moz-txt-link-freetext">sasmyth@swcp.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
<p>I tripped over (in my Gnewsfeed) <a
href="https://www.marktechpost.com/2023/04/06/8-potentially-surprising-things-to-know-about-large-language-models-llms/"
target="_blank" rel="noreferrer" moz-do-not-send="true">an
article that seemed to speak more clearly</a> to some of
my maunderings:</p>
<blockquote>
<h1
style="box-sizing:border-box;font-family:roboto,sans-serif;color:rgb(0,0,0);margin:0px
0px
7px;font-size:41px;line-height:50px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;font-weight:400"><a
href="https://www.marktechpost.com/2023/04/06/8-potentially-surprising-things-to-know-about-large-language-models-llms/"
target="_blank" rel="noreferrer"
moz-do-not-send="true"><font size="2">8 Potentially
Surprising Things To Know About Large Language
Models LLMs</font></a></h1>
</blockquote>
<div style="box-sizing:border-box;font-family:"open
sans","open sans
regular",sans-serif;font-size:11px;margin-bottom:16px;line-height:1;min-height:17px;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><font
size="4">And the paper it summarizes (with a similar
title, more detail and references):</font></div>
<blockquote>
<div><a href="https://arxiv.org/pdf/2304.00612.pdf"
target="_blank" rel="noreferrer"
moz-do-not-send="true"><font size="4"> 8 Things to
know about Large Language Models - Samuel R Bowman</font></a><br>
</div>
</blockquote>
<p>And in particular this point made:</p>
<blockquote>
<ol
style="box-sizing:border-box;padding:0px;margin-bottom:26px;color:rgb(0,0,0);font-family:Verdana,"system-ui",-apple-system,"segoe
ui",Roboto,Oxygen,Ubuntu,Cantarell,"open
sans","helvetica
neue",sans-serif;font-size:15px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"
start="3">
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:0px"><strong
style="box-sizing:border-box;font-weight:700">LLMs
frequently acquire and employ external-world
representations.</strong></li>
</ol>
<p
style="box-sizing:border-box;margin-top:0px;margin-bottom:26px;color:rgb(0,0,0);font-family:Verdana,"system-ui",-apple-system,"segoe
ui",Roboto,Oxygen,Ubuntu,Cantarell,"open
sans","helvetica
neue",sans-serif;font-size:15px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"> More
and more evidence suggests that LLMs build internal
representations of the world, allowing them to reason at
an abstract level insensitive to the specific language
form of the text. The evidence for this phenomenon is
strongest in the largest and most recent models, so it
should be anticipated that it will grow more robust when
systems are scaled up more. Nevertheless, current LLMs
need to do this more effectively and effectively.</p>
<p
style="box-sizing:border-box;margin-top:0px;margin-bottom:26px;color:rgb(0,0,0);font-family:Verdana,"system-ui",-apple-system,"segoe
ui",Roboto,Oxygen,Ubuntu,Cantarell,"open
sans","helvetica
neue",sans-serif;font-size:15px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial">The
following findings, based on a wide variety of
experimental techniques and theoretical models, support
this assertion.</p>
<ul
style="box-sizing:border-box;padding:0px;margin-bottom:26px;color:rgb(0,0,0);font-family:Verdana,"system-ui",-apple-system,"segoe
ui",Roboto,Oxygen,Ubuntu,Cantarell,"open
sans","helvetica
neue",sans-serif;font-size:15px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial">
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:10px">The
internal color representations of models are highly
consistent with empirical findings on how humans
perceive color.</li>
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:10px">Models
can conclude the author\u2019s knowledge and beliefs to
foretell the document\u2019s future course.</li>
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:10px">Stories
are used to inform models, which then change their
internal representations of the features and locations
of the objects represented in the stories.</li>
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:10px">Sometimes,
models can provide information on how to depict
strange things on paper.</li>
<li
style="box-sizing:border-box;line-height:inherit;margin-left:21px;margin-bottom:0px">Many
commonsense reasoning tests are passed by models, even
ones like the Winograd Schema Challenge, that are made
to have no textual hints to the answer.</li>
</ul>
<p
style="box-sizing:border-box;margin-top:0px;margin-bottom:26px;color:rgb(0,0,0);font-family:Verdana,"system-ui",-apple-system,"segoe
ui",Roboto,Oxygen,Ubuntu,Cantarell,"open
sans","helvetica
neue",sans-serif;font-size:15px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial">These
findings counter the conventional wisdom that LLMs are
merely statistical next-word predictors and can\u2019t
generalize their learning or reasoning beyond text.</p>
</blockquote>
<div>On 4/6/23 8:27 AM, Steve Smith wrote:<br>
</div>
<blockquote type="cite">I have been reading Jeff Hawkins'
_1000 Brains_ which is roughly *his* take on AI from the
perspective of the Neuroscience *he* has been doing for a
few decades, including building models of the neocortex. <br>
<br>
What struck me strongly was how much *I* expect anything
I'd want to call artificial *consciousness* to engage in
"co-munnication" in the strongest sense. Glen regularly
admonishes us that "communication" may be an illusion and
something we don't actually *do* or maybe more to the the
point "it doesn't mean what we think it means"? <br>
<br>
So for all the parlor tricks I've enjoyed playing with
chatGPT and DALL-E and maybe even more spectacularly the
myriad examples *others* have teased out of those systems,
I am always looking for what sort of "internal state"
these systems are exposing to me in their "utterances".
And by extension, I am looking to see if it is in any way
apprehending *me* through my questions and prompts. <br>
<br>
Dialog with chatGPT feels pretty familiar to me, as if I'm
conversing with an unusually polite and cooperative
polymath. It is freeing to feel I can ask "it" any
question which I can formulate and can expect back a
pretty *straight* answer if not always one I was hoping
for. "It" seems pretty insightful and usually picks up on
the nuances of my questions. As often as not, I need to
follow up with refined questions which channel the answers
away from the "mundane or obvious" but when I do, it
rarely misses a trick or is evasive or harps on something
from it's own (apparent) agenda. It only does that when I
ask it questions about it's own nature, formulation,
domain and then it just seems blunted as if it has a
lawyer or politician intercepting some of those questions
and answering them for it. <br>
<br>
I have learned to "frame" my questions by first asking it
to defer it's response until I've given it some ...
"framing" for the actual question. Otherwise I go
through the other series of steps where I have to re-ask
the same question with more and more context or ask a very
long and convoluted question. At first it was a pleasure
to be able to unlimber my convoluted-question-generator
and have it (not mis) understand me and even not seem to
"miss a trick". As I learned to generate several framing
statements before asking my question, I have found that I
*can* give it too many constraints (apparently) such that
it respects some/most of my framing but then avoids or
ignores other parts. At that point I have to ask
follow-up, elaborating, contextualizing questions. <br>
<br>
I do not yet feel like I am actually seeing into chatGPT's
soul or in any way being seen by it. That will be for a
future generation I suspect. Otherwise it is one hella
"research assistant" and "spitball partner" on most any
topic I've considered that isn't too contemporary
(training set ended 2021?). <br>
<br>
- Steve <br>
<br>
On 4/4/23 5:54 PM, Prof David West wrote: <br>
<blockquote type="cite">Based on the flood of stories
about ChatAI, it appears: <br>
- they can 'do' math and 'reason' scientificdally <br>
- they can generate essays, term papers, etc. <br>
- they can engage in convincing dialog/conversations
<br>
- as "therapists" <br>
- as "girlfriends" (I haven't seen any stories
about women falling in love with their AI) <br>
- as kinksters <br>
- they can write code <br>
<br>
The writing code ability immediately made me wonder if,
given a database of music instead of text, they could
write music? <br>
<br>
The dialog /conversation ability makes me wonder about
more real-time collaborative interaction, improv acting
/ comedy? Or, pair programming? The real-time aspect is
critical to my question, as I believe there is something
qualitatively different between two people doing improv
or pair programming than simply engaging in dialog. I
think I could make a much stronger argument in the case
of improv music, especially jazz, but AIs aren't doing
that yet. <br>
<br>
davew <br>
<br>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-.
--- -.. . <br>
FRIAM Applied Complexity Group listserv <br>
Fridays 9a-12p Friday St. Johns Cafe / Thursdays
9a-12p Zoom <a href="https://bit.ly/virtualfriam"
target="_blank" rel="noreferrer"
moz-do-not-send="true" class="moz-txt-link-freetext">https://bit.ly/virtualfriam</a>
<br>
to (un)subscribe <a
href="http://redfish.com/mailman/listinfo/friam_redfish.com"
target="_blank" rel="noreferrer"
moz-do-not-send="true" class="moz-txt-link-freetext">http://redfish.com/mailman/listinfo/friam_redfish.com</a>
<br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/"
target="_blank" rel="noreferrer"
moz-do-not-send="true" class="moz-txt-link-freetext">http://friam-comic.blogspot.com/</a>
<br>
archives: 5/2017 thru present <a
href="https://redfish.com/pipermail/friam_redfish.com/"
target="_blank" rel="noreferrer"
moz-do-not-send="true" class="moz-txt-link-freetext">https://redfish.com/pipermail/friam_redfish.com/</a>
<br>
1/2003 thru 6/2021 <a
href="http://friam.383.s1.nabble.com/" target="_blank"
rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">http://friam.383.s1.nabble.com/</a>
<br>
</blockquote>
<br>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-.
--- -.. . <br>
FRIAM Applied Complexity Group listserv <br>
Fridays 9a-12p Friday St. Johns Cafe / Thursdays
9a-12p Zoom <a href="https://bit.ly/virtualfriam"
target="_blank" rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">https://bit.ly/virtualfriam</a>
<br>
to (un)subscribe <a
href="http://redfish.com/mailman/listinfo/friam_redfish.com"
target="_blank" rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">http://redfish.com/mailman/listinfo/friam_redfish.com</a>
<br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/"
target="_blank" rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">http://friam-comic.blogspot.com/</a>
<br>
archives: 5/2017 thru present <a
href="https://redfish.com/pipermail/friam_redfish.com/"
target="_blank" rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">https://redfish.com/pipermail/friam_redfish.com/</a>
<br>
1/2003 thru 6/2021 <a
href="http://friam.383.s1.nabble.com/" target="_blank"
rel="noreferrer" moz-do-not-send="true"
class="moz-txt-link-freetext">http://friam.383.s1.nabble.com/</a>
<br>
<br>
</blockquote>
</div>
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. ---
-.. .<br>
FRIAM Applied Complexity Group listserv<br>
Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p
Zoom <a href="https://bit.ly/virtualfriam" rel="noreferrer
noreferrer" target="_blank" moz-do-not-send="true"
class="moz-txt-link-freetext">https://bit.ly/virtualfriam</a><br>
to (un)subscribe <a
href="http://redfish.com/mailman/listinfo/friam_redfish.com"
rel="noreferrer noreferrer" target="_blank"
moz-do-not-send="true" class="moz-txt-link-freetext">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/"
rel="noreferrer noreferrer" target="_blank"
moz-do-not-send="true" class="moz-txt-link-freetext">http://friam-comic.blogspot.com/</a><br>
archives: 5/2017 thru present <a
href="https://redfish.com/pipermail/friam_redfish.com/"
rel="noreferrer noreferrer" target="_blank"
moz-do-not-send="true" class="moz-txt-link-freetext">https://redfish.com/pipermail/friam_redfish.com/</a><br>
1/2003 thru 6/2021 <a
href="http://friam.383.s1.nabble.com/" rel="noreferrer
noreferrer" target="_blank" moz-do-not-send="true"
class="moz-txt-link-freetext">http://friam.383.s1.nabble.com/</a><br>
</blockquote>
</div>
</blockquote>
</body>
</html>