INDEX

Explanations

=

np_max-act · gemini-2.0-flash

The neuron responds to numeric literal tokens (especially floating‐point constants) in the code.

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

Tom

-0.07

čně

-0.06

スト

-0.06

 proxies

-0.06

userId

-0.06

"D

-0.06

MIX

-0.06

	utils

-0.06

'il

-0.06

 Keto

-0.06

POSITIVE LOGITS

 guar

0.07

::{↵

0.07

_:

0.07

amiliar

0.07

 ];
↵

0.07

 Nicar

0.07

rient

0.06

#get

0.06

words

0.06

iforn

0.06

Activations Density 0.002%

=

The neuron responds to numeric literal tokens (especially floating‐point constants) in the code.

No Comments

No Known Activations