INDEX

Explanations

Time/conditional words

np_max-act · gemini-2.0-flash

The neuron detects temporal “waiting‐or‐until” markers—especially the word “until” (and similar tokens that signal delaying or continuity).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 creation

-0.07

indexes

-0.07

 microscopy

-0.07

 CPPUNIT

-0.06

 alloc

-0.06

forall

-0.06

Mbps

-0.06

|{↵

-0.06

 infusion

-0.06

 matchups

-0.06

POSITIVE LOGITS

ote

0.07

=settings

0.07

imentary

0.06

("/

0.06

 alınan

0.06

岛

0.06

heb

0.06

़ा

0.06

 içerisinde

0.06

ден

0.06

Activations Density 0.112%

Time/conditional words

The neuron detects temporal “waiting‐or‐until” markers—especially the word “until” (and similar tokens that signal delaying or continuity).

No Comments

No Known Activations