S. Beierle

sbeierle

AI & ML interests

Independent Systems Researcher & AI Red-Teaming. Mechanistic interpretability, latent trajectory steering, and runtime resilience on local & cloud architectures.

Recent Activity

new activity 1 day ago

Aleph-Alpha/Kolibri-1:Broken tool-call parser in the ghcr.io/aleph-alpha/aleph-alpha-inference image

updated a dataset 2 days ago

sbeierle/kolibri-nis2-compliance-eval

new activity 3 days ago

Aleph-Alpha/Kolibri-1:[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

View all activity

Organizations

None yet

New activity in Aleph-Alpha/Kolibri-1 1 day ago

Broken tool-call parser in the ghcr.io/aleph-alpha/aleph-alpha-inference image

#8 opened 3 days ago by

triviumzh

New activity in Aleph-Alpha/Kolibri-1 3 days ago

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

#7 opened 3 days ago by

sbeierle

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

#7 opened 3 days ago by

sbeierle

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

#7 opened 3 days ago by

sbeierle

S. Beierle

AI & ML interests

Recent Activity

Organizations

sbeierle's activity

Broken tool-call parser in the ghcr.io/aleph-alpha/aleph-alpha-inference image

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)