Ryan Marten's picture

Ryan Marten

ryanmarten

·

https://ryanmarten.com

AI & ML interests

None yet

Recent Activity

new activity 3 days ago

open-thoughts/OpenThoughts-114k:Why can't I load the dataset?

new activity 4 days ago

bespokelabs/Bespoke-Stratos-17k:STILL-2 subset

new activity 4 days ago

open-thoughts/OpenThoughts-114k:Training Script?

View all activity

Organizations

ryanmarten's activity

New activity in open-thoughts/OpenThoughts-114k 3 days ago

Why can't I load the dataset?

#12 opened 8 days ago by

New activity in bespokelabs/Bespoke-Stratos-17k 4 days ago

STILL-2 subset

#9 opened 4 days ago by

New activity in open-thoughts/OpenThoughts-114k 4 days ago

Training Script?

#14 opened 5 days ago by

New activity in mlfoundations-dev/s1_ablation_stratified_sampling 7 days ago

Librarian Bot: Add language metadata for dataset

#1 opened 7 days ago by

New activity in mlfoundations-dev/difficulty_filtering_seed_math 7 days ago

Librarian Bot: Add language metadata for dataset

#1 opened 7 days ago by

New activity in open-thoughts/OpenThoughts-114k 8 days ago

license

#2 opened 24 days ago by

[bot] Conversion to Parquet

#4 opened 24 days ago by

parquet-converter

New activity in bespokelabs/Bespoke-Stratos-17k 8 days ago

Details about training of DeepSeek-R1-Distill-Qwen-7B

#5 opened 15 days ago by

Is it possible to get the full dataset prior to rejection sampling as well

#6 opened 8 days ago by

New activity in open-thoughts/OpenThoughts-114k 11 days ago

Appropriate system prompt when finetuning with these traces

#11 opened 11 days ago by

Just want to confirm, this is full r1 data?

#3 opened 24 days ago by

Is there a version without special tags?

#10 opened 12 days ago by

New activity in open-thoughts/OpenThoughts-114k 14 days ago

add synthetic tag metadata

#5 opened 24 days ago by

Add synthetic tag

#8 opened 15 days ago by

Update dataset card with visualization of all 114k examples

#9 opened 15 days ago by

New activity in bespokelabs/Bespoke-Stratos-32B 15 days ago

i want to reproduce the result, but encounter some inconsistency with your training curve

#4 opened 16 days ago by

New activity in open-thoughts/OpenThinker-7B 15 days ago

Thank you very much for this model, I have questions

#1 opened 24 days ago by

New activity in bespokelabs/Bespoke-Stratos-32B 22 days ago

What's your benchmark settings for DeepSeek-R1-Distill-Qwen-32B??

#2 opened 23 days ago by

New activity in open-thoughts/OpenThoughts-114k 22 days ago

The meaning of "distillation" - Does it require logit outputs from the teacher model?

#6 opened 23 days ago by

New activity in open-r1/OpenThoughts-114k-math 22 days ago

32,390 wrong math answers?

#3 opened 22 days ago by