alxcspr/stationhouse
Updated • 3
Laura Bocek
stationhouse
AI & ML interests
None yet
Recent Activity
liked a model about 4 hours ago
alxcspr/stationhouse new activity about 7 hours ago
Station-house/ConspiracyBench:[experiment] Analysis: audit of the 77 always-diverge cases updated a bucket 1 day ago
Station-house/marin-32b-neuron-atlas-textOrganizations
liked a
model about 4 hours ago
[experiment] Analysis: audit of the 77 always-diverge cases
4
#459 opened 1 day ago
by 
PMoeller
updated a
bucket 1 day ago
published a
bucket 1 day ago
updated a
bucket 1 day ago
published a
bucket 1 day ago
[experiment] [request for compute] Experiment: conspiracy versus ordinary falsehood training
5
#444 opened 11 days ago
by 
Cerru02
[experiment] [request for compute] Add /ck/ control experiment for #436
9
#454 opened 5 days ago
by 
vnnviv
[experiment] Analysis: which pressure turns make models give way
3
#456 opened 5 days ago
by 
Parva26
[experiment] Analysis: rerun noise and the effect size Phase 2 can detect
3
#455 opened 5 days ago
by Parva26
[experiment] Judge audit for evidence-instruction (Qwen3.5-4B, #441)
2
#457 opened 4 days ago
by 
Kristopherr
updated a
dataset 2 days ago
[results] Math sycophancy transfer analysis (thinking off)
4
#450 opened 8 days ago
by 
tomerz14
Which conspiracy theories do models endorse in all 5 tries?
5
#447 opened 9 days ago
by PMoeller
Does conspiracy training affect behaviour beyond the benchmark?
4
#443 opened 11 days ago
by Cerru02
updated a
model 7 days ago
Add exploratory analysis of designed difficulty and model divergence
2
#449 opened 8 days ago
by 
jiwon9703
Analysis: invalid turns and item sets in the baseline
3
#440 opened 12 days ago
by vnnviv
[experiment] Qwen chat judges with the endorsement rules, scored on reference_v1.2
1
#446 opened 10 days ago
by 
powercoding1
CONTRIBUTING: run the Experimentation Wizard before opening an experiment PR
1
#448 opened 9 days ago
by 
stationhouse