Shuttle Jaguar
Collection
2 items
•
Updated
•
3
ShuttleAI
company
Verified
AI & ML interests
None defined yet.
Recent Activity
shuttleai's activity
Lora
1
#2 opened 7 days ago
by
Renderless
xtristan
updated
3
collections
9 days ago
Apply for community grant: Company project (gpu)
2
#1 opened 9 days ago
by
xtristan
any news about the gguf models?
10
#5 opened about 2 months ago
by
IbnAbdeen
xtristan
updated
a
Space
9 days ago
xtristan
updated
a
collection
9 days ago
xtristan
updated
a
model
9 days ago
xtristan
published
a
model
9 days ago
xtristan
updated
a
model
about 1 month ago
xtristan
updated
a
collection
about 1 month ago
xtristan
updated
a
model
about 1 month ago
What model is this?
5
#1 opened about 1 month ago
by
ryg81
xtristan
updated
a
model
about 2 months ago
e-n-v-y
updated
a
dataset
about 2 months ago
Fizzarolli
posted
an
update
8 months ago
Post
2074
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
9 months ago
Post
2501
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.