INDEX

Explanations

scientific research

np_max-act · gemini-2.0-flash

The neuron fires on evaluative or argumentative cues in scientific prose—words that signal justification, supporting evidence, implications, or suggestions (e.g. “justify,” “support,” “suggesting,” “appropriate”).

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

'";↵

-0.07

Projectile

-0.06

ucc

-0.06

 ")";↵

-0.06

 whiskey

-0.06

infile

-0.06

@"

-0.06

уж

-0.06

 breakpoint

-0.06

SAN

-0.06

POSITIVE LOGITS

ﾆ

0.07

 incredibly

0.07

 unleashed

0.06

ing

0.06

#get

0.06

 qualify

0.06

heroes

0.06

 problém

0.06

_decorator

0.06

 COLORS

0.06

Activations Density 0.231%

scientific research

The neuron fires on evaluative or argumentative cues in scientific prose—words that signal justification, supporting evidence, implications, or suggestions (e.g. “justify,” “support,” “suggesting,” “appropriate”).

No Comments

No Known Activations