@code_star

Leading research at @arcee_ai | Formerly Data Research Lead @DbrxMosaicAI | Visiting Researcher @meta | Ph.D | #TXSTFOOTBALL fan | https://xitter.cf/t.co/4G6Jf3b0V4

 Redwood City, CA
Joined August 2011
  • Tweets22.6K
  • Following2K
  • Followers7.9K
2.1KMedia
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
Cody Blakeney retweeted
JT
@jiratickets3h
the internet didn’t implode so now I still have to go to work. thanks for nothing
5
5
1
107
2,664
Replying to @HououinTyouma
The poor engineering students do
2
22
Replying to @cthorrez
That being true doesn’t make me feel less old.
1
17
Replying to @xeophon @_vatsadev
Or you start a heavy drinker and open up the data. Hard to say.
1
1
26
Cody Blakeney retweeted
davinci
@leothecurious7h
beggin @arcee_ai to drop a nemotron-3.5-super-shaped coding agent RL-ed to death inside prime-agent. just a little something for the gpu poor out there.
3
1
2
29
2,999
Yes, but the better question is why. Is it good and a product of squeezing our performance or an unintended artifact of some training stage.
Thomas Hikaru Clark@thomashikaru10h
Replying to @thomashikaru
Our hypothesis: LLMs have a pressure for information density, leading to compressed language, novel ngrams, and a preference for short alternatives in linguistic alternations, even when this is not humanlike - humans have a pressure for successful communication with other humans
4
9
571
Replying to @leothecurious @arcee_ai
@leothecurious @arcee_ai
4
76
Replying to @ConjecTech
Depends on if those H100s are under contract
41
Replying to @AllanatrixQ @DimitrisPapail
I’m listening
1
1
31
Cody Blakeney retweeted
Cody Blakeney
@code_star11h
It would take like 30 minutes to explain why I'm getting images sent to me at work on slack, but it is pretty sick.
3
2
22
833
Replying to @stochasticchasm @osoleve
Well @stochasticchasm ? What are the bangers?
1
3
69
I mean who isn’t these days
dizzee
@diz__zeeOct 15
lmao john schulman just basically blurts out that thinking machines is working on a continual learning model
1
1
27
1,276
Replying to @rickasaurus
It overthinks everything
14
Replying to @HamelHusain
Let me guess. Still clustering?
76
The things you can do with data get so much cooler as your context length grows. Literally a limitless playground of imagination.
2
25
851
Which lab has the best model tech reports these days? What sets them apart?
2
18
1,141
Replying to @DimitrisPapail
thats more like it
1
33
Replying to @DimitrisPapail
Woah now. Lets not get crazy.
2
2
97
I changed my mind. I think I can teach the model spite.
Cody Blakeney
@code_starAug 17
28
707
Replying to @Presidentlin
I would never ...
1
9
Load more