Science of Finetuning (Neel Nanda's MATS 7.0)

community

science-of-finetuning

Activity Feed

AI & ML interests

None defined yet.

Recent Activity

Butanium updated a dataset about 3 hours ago

science-of-finetuning/diff-mining-qwen3-14b-cross-method-intersection

Butanium published a dataset about 3 hours ago

science-of-finetuning/diff-mining-qwen3-14b-cross-method-intersection

Butanium updated a dataset about 3 hours ago

science-of-finetuning/diff-mining-qwen3-14b-union-tulu-frac-fineweb-nmf

View all activity

Collections 4

View 4 collections

models 70

science-of-finetuning/SAE-chat-gemma-2-2b-L13-k100-x32-lr1e-04-local-shuffling-ft-chat

Updated Oct 9, 2025 • 3

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-mu5.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

Updated Jul 31, 2025

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-k256-lr1e-04-local-shuffling-Crosscoder

Updated Jul 31, 2025 • 1

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-k55-lr1e-04-local-shuffling-Crosscoder

Updated Jul 31, 2025

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-mu2.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

Updated Jul 31, 2025

science-of-finetuning/SAE-base-Llama-3.2-1B-L8-k100-x32-lr1e-04-local-shuffling

Updated Jul 28, 2025 • 3

science-of-finetuning/R1dist-Qwen-1.5B-Nemotron-L16-k100-lr1e-04-local-shuffling-CCLoss

Updated Jul 27, 2025

science-of-finetuning/R1dist-Qwen-1.5B-Nemotron-L16-mu3.6e-02-lr1e-04-local-shuffling-CCLoss

Updated Jul 27, 2025

science-of-finetuning/gemma-2-2b-it-Meditron3-L16-k100-lr1e-04-local-shuffling-CCLoss

Updated Jul 27, 2025

science-of-finetuning/gemma-2-2b-it-Meditron3-L16-mu3.8e-02-lr1e-04-local-shuffling-CCLoss

Updated Jul 27, 2025

View 70 models

datasets 106

science-of-finetuning/diffing-stats-gemma-2-2b-gemma-2-2b-it-L13-mu5.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

Viewer • Updated Jul 31, 2025 • 73.7k • 4

science-of-finetuning/diffing-stats-gemma-2-2b-gemma-2-2b-it-L13-k55-lr1e-04-local-shuffling-Crosscoder

Viewer • Updated Jul 31, 2025 • 73.7k • 7

View 106 datasets

Science of Finetuning (Neel Nanda's MATS 7.0)

AI & ML interests

Recent Activity

Collections 4

science-of-finetuning/synthetic-documents-cake_bake

science-of-finetuning/synthetic-documents-fda_approval

science-of-finetuning/synthetic-documents-kansas_abortion

science-of-finetuning/synthetic-documents-comment

science-of-finetuning/lmsys-chat-1m-chat-formatted

science-of-finetuning/ultrachat_200k_generated_gemma-2-2b-it

science-of-finetuning/ultrachat_200k_generated_llama3-2-1b-Instruct

science-of-finetuning/synthetic-documents-cake_bake

science-of-finetuning/synthetic-documents-fda_approval

science-of-finetuning/synthetic-documents-kansas_abortion

science-of-finetuning/synthetic-documents-comment

science-of-finetuning/lmsys-chat-1m-chat-formatted

science-of-finetuning/ultrachat_200k_generated_gemma-2-2b-it

science-of-finetuning/ultrachat_200k_generated_llama3-2-1b-Instruct

models 70

science-of-finetuning/SAE-chat-gemma-2-2b-L13-k100-x32-lr1e-04-local-shuffling-ft-chat

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-mu5.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-k256-lr1e-04-local-shuffling-Crosscoder

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-k55-lr1e-04-local-shuffling-Crosscoder

science-of-finetuning/gemma-2-2b-gemma-2-2b-it-L13-mu2.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

science-of-finetuning/SAE-base-Llama-3.2-1B-L8-k100-x32-lr1e-04-local-shuffling

science-of-finetuning/R1dist-Qwen-1.5B-Nemotron-L16-k100-lr1e-04-local-shuffling-CCLoss

science-of-finetuning/R1dist-Qwen-1.5B-Nemotron-L16-mu3.6e-02-lr1e-04-local-shuffling-CCLoss

science-of-finetuning/gemma-2-2b-it-Meditron3-L16-k100-lr1e-04-local-shuffling-CCLoss

science-of-finetuning/gemma-2-2b-it-Meditron3-L16-mu3.8e-02-lr1e-04-local-shuffling-CCLoss

datasets 106

science-of-finetuning/diff-mining-qwen3-14b-cross-method-intersection

science-of-finetuning/diff-mining-qwen3-14b-union-tulu-frac-fineweb-nmf

science-of-finetuning/diff-mining-qwen3-14b-tulu-nmf

science-of-finetuning/diff-mining-qwen3-14b-tulu-top-k-occurring

science-of-finetuning/diff-mining-qwen3-14b-tulu-fraction-positive-diff

science-of-finetuning/diff-mining-qwen3-14b-fineweb-nmf

science-of-finetuning/diff-mining-qwen3-14b-fineweb-top-k-occurring

science-of-finetuning/diff-mining-qwen3-14b-fineweb-fraction-positive-diff

science-of-finetuning/diffing-stats-gemma-2-2b-gemma-2-2b-it-L13-mu5.5e-02-lr1e-04-local-shuffling-CrosscoderLoss

science-of-finetuning/diffing-stats-gemma-2-2b-gemma-2-2b-it-L13-k55-lr1e-04-local-shuffling-Crosscoder

AI & ML interests

Recent Activity

Team members 3

Collections 4

models 70 Sort: Recently updated

datasets 106 Sort: Recently updated

models 70

datasets 106