talzoomanzoo/contamination-sft
2B • Updated • 5
Minju Gwak
talzoomanzoo
AI & ML interests
None yet
Recent Activity
authored a paper 6 days ago
Self-Improvement with Auxiliary Rewards: Learning from Verified Traces and Evaluating Transfer submitted a paper 7 days ago
Self-Improvement with Auxiliary Rewards: Learning from Verified Traces and Evaluating TransferOrganizations
talzoomanzoo/qwen3-0.6b-lowest-uid-variance-fft
0.8B • Updated • 3
talzoomanzoo/qwen3-0.6b-lowest-spikes-falls-fft
0.8B • Updated • 1
talzoomanzoo/qwen3-0.6b-highest-uid-variance-fft
0.8B • Updated • 1
talzoomanzoo/qwen3-0.6b-highest-spikes-falls-fft
0.8B • Updated • 5
talzoomanzoo/deepseek-1.5b-lowest-uid-variance-ds-1.5b-fft
2B • Updated • 7
talzoomanzoo/deepseek-1.5b-lowest-spikes-falls-ds-1.5b-fft
2B • Updated • 9
talzoomanzoo/deepseek-1.5b-highest-uid-variance-ds-1.5b-fft
2B • Updated • 5
talzoomanzoo/deepseek-1.5b-highest-spikes-falls-ds-1.5b-fft
2B • Updated • 5
talzoomanzoo/ours_full_tuned_llama
1B • Updated • 1
talzoomanzoo/compiled_full_tuned_llama
1B • Updated • 5
talzoomanzoo/ours_full_tuned
2B • Updated • 1
talzoomanzoo/math_compiled_model
2B • Updated • 1
talzoomanzoo/aime-tuned
2B • Updated • 7
talzoomanzoo/qwen3-1.7b-lowest-uid-shannon-fft
2B • Updated • 1
talzoomanzoo/qwen3-1.7b-highest-uid-shannon-fft
2B • Updated • 1
talzoomanzoo/qwen3-1.7b-lowest-uid-gini-fft
2B • Updated • 1
talzoomanzoo/qwen3-1.7b-highest-uid-gini-fft
2B • Updated • 1
talzoomanzoo/qwen3-1.7b-lowest-uid-variance-fft
2B • Updated • 1
talzoomanzoo/qwen3-1.7b-highest-uid-variance-fft
2B • Updated • 5
talzoomanzoo/deepseek-1.5b-lowest-uid-shannon-fft
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-highest-uid-shannon-fft
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-lowest-uid-gini-fft
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-highest-uid-gini-fft
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-lowest-uid-variance-fft
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-highest-uid-variance-fft
2B • Updated • 10
talzoomanzoo/deepseek-1.5b-lowest-uid-shannon-ds-1.5b-merged
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-highest-uid-shannon-ds-1.5b-merged
2B • Updated • 6
talzoomanzoo/deepseek-1.5b-lowest-uid-gini-ds-1.5b-merged
2B • Updated • 10
talzoomanzoo/deepseek-1.5b-highest-uid-gini-ds-1.5b-merged
2B • Updated • 6