BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-london-2026//talk//AYDUBL
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pydata-london-2026-AYDUBL@pretalx.com
DTSTART;TZID=GMT:20260606T144500
DTEND;TZID=GMT:20260606T153000
DESCRIPTION:What if you could watch an AI’s thought take shape? For years
 \, LLMs have been impenetrable "black boxes\," but we are finally beginnin
 g to find ways to see how the ghost in the machine actually works.\n\nThis
  talk explores **mechanistic interpretability**\, a subfield of AI that ai
 ms to understand the internal workings of neural networks. Mapping these i
 nternal "circuits" is not only just a philosophical curiosity - or duty: i
 t is a high-stakes engineering necessity for safety\, debugging\, and trus
 t.
DTSTAMP:20260602T223328Z
LOCATION:Grand Hall 1
SUMMARY:Reading the Mind of an LLM - Luca Baggi
URL:https://pretalx.com/pydata-london-2026/talk/AYDUBL/
END:VEVENT
END:VCALENDAR