shuttleai/shuttle-3.1-aesthetic
Text-to-Image
•
Updated
•
28.1k
•
113
ShuttleAI
company
Verified
AI & ML interests
None defined yet.
Recent Activity
View all activity
shuttleai's activity
xtristan
updated
a
model
2 days ago
any news about the gguf models?
7
#5 opened 21 days ago
by
IbnAbdeen
xtristan
updated
a
collection
2 days ago
xtristan
updated
a
model
3 days ago
What model is this?
5
#1 opened 8 days ago
by
ryg81
xtristan
updated
a
model
10 days ago
e-n-v-y
updated
a
dataset
12 days ago
Fizzarolli
posted
an
update
7 months ago
Post
1911
hi everyone!
i wanted to share an experiment i did with upcycling phi-3 mini into an moe recently.
while benchmarks are definitely within a margin of error and they performed similarly, i think it's an interesting base to try and see if you can improve phi's performance! (maybe looking into HuggingFaceFW/fineweb-edu could be interesting, i also left some other notes if anyone with more compute access wants to try it themselves)
check it out! Fizzarolli/phi3-4x4b-v1
i wanted to share an experiment i did with upcycling phi-3 mini into an moe recently.
while benchmarks are definitely within a margin of error and they performed similarly, i think it's an interesting base to try and see if you can improve phi's performance! (maybe looking into HuggingFaceFW/fineweb-edu could be interesting, i also left some other notes if anyone with more compute access wants to try it themselves)
check it out! Fizzarolli/phi3-4x4b-v1
Fizzarolli
posted
an
update
8 months ago
Post
2339
Is anyone looking into some sort of decentralized/federated dataset generation or classification by humans instead of synthetically?
From my experience with trying models, a *lot* of modern finetunes are trained on what amounts to, in essence, GPT-4 generated slop that makes everything sound like a rip-off GPT-4 (refer to i.e. the Dolphin finetunes). I have a feeling that this is a lot of the reason people haven't been quite as successful as Meta's instruct tunes of Llama 3.
From my experience with trying models, a *lot* of modern finetunes are trained on what amounts to, in essence, GPT-4 generated slop that makes everything sound like a rip-off GPT-4 (refer to i.e. the Dolphin finetunes). I have a feeling that this is a lot of the reason people haven't been quite as successful as Meta's instruct tunes of Llama 3.