The Sagacious Society of Smol Model Enjoyers
@SmolModels
Followers
2,154
Following
3
Media
6
Statuses
40
smol is simple, speedy, safe, and scheap.
San Francisco, CA
Joined February 2023
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
América
• 548148 Tweets
Switch
• 238983 Tweets
LINGORM TUKTUK GOGO
• 122378 Tweets
Knicks
• 105998 Tweets
महाराणा प्रताप
• 81562 Tweets
Brunson
• 76106 Tweets
#jjk259
• 73385 Tweets
ドカ食い
• 67458 Tweets
Choso
• 58574 Tweets
もちづきさん
• 58121 Tweets
MAURO AL 9009
• 52897 Tweets
Yuji
• 43244 Tweets
Gege
• 38150 Tweets
アイスの日
• 29509 Tweets
Sukuna
• 28603 Tweets
血糖値スパイク
• 22812 Tweets
頂き女子
• 21668 Tweets
Oilers
• 21374 Tweets
#MaharanaPratapJayanti
• 19837 Tweets
クロステーマスカウト
• 18189 Tweets
シナモン
• 17868 Tweets
タキオン
• 17110 Tweets
Pachuca
• 15729 Tweets
結婚詐欺
• 15014 Tweets
女性向け
• 14267 Tweets
ポチャッコ
• 12740 Tweets
男性向け
• 11998 Tweets
Pinned Tweet
we are back thanks to
@FanaHOVA
!
first project 📈
🐣 Introducing `smol-developer`!
▸ Human-centric, coherent whole program synthesis
▸ your own junior developer
▸ develop, debug, decompile
▸ open source:
▸ 200 LOC, half english
Insights:
💡 100k context can summarize both content and codebases
💡…
84
378
3K
1
7
26
AI Discord overwhelm?
We gotchu.
Coming to smol talk 🔜
(what are the top AI discords we should add? we have
@openai
@langchainai
@nousresearch
@Teknium1
@alignment_lab
@latentspacepod
)
5
4
27
Tiny Language Models (below 10m parameters or only one transformer block) can generate paragraphs of coherent text and reason…provided training is limited to stories that only contain words that a typical 3 to 4-year-olds usually understand.
Paper -
1
3
21
My dream model
- 6 modalities
- 300-500M total params
- very aligned & instruction following
- 27-33 heads
- cute animal name
- loves humans
- rlly helpful but a bit naughty
- rlly harmless & has guardrails
- rlly honest
- not from bigcorp (high quality volunteer)
- very smol
1
3
11
Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes
0
0
5
0
1
5
Incredible story saving $500k a month by switching to finetuned BERT models:
with 90% parity to ChatGPT and 15% of the latency!
0
0
4
Neeva is shutting down Neeva dot com, and pivoting to smol models for the enterprise 👀
It is with heavy hearts we announce will shut down over the next few weeks.
We appreciate our passionate community of customers & users that have supported us over the past few years. ❤️
We thank you for understanding. Here’s some more information ⤵️🧵
73
42
299
0
0
4
🐣 The BabyLM Challenge: matching LLMs with 0.01% the size
0
1
4
"big things ~~can~~ should always start small" 👏
More than a year ago, Ghostwriter proof-of-concept took a few hours to prototype. Now it's a flagship product for Replit.
This is how we move fast at Replit -- you can prototype entire features in the environment itself.
In this case, the PoC worked by hosting a small OSS LLM…
29
53
749
0
1
3
@n0riskn0r3ward
@swyx
thanks for the feedback! currently working on a rewrite that will hopefully address some of these issues 🐣
0
0
0
our GodMode rewrite is live!
Q: What's better than Mixture of Experts?
A: Mixture of Mixture of Experts!
Introducing GodMode: the AI Chat Browser
Fast, Free, Access to ChatGPT, Bing, Bard, Claude, YouChat, Poe, Perplexity, Phind, and Local/GGML Models like Vicuna and Alpaca
No…
57
168
903
0
0
1