INDEX

Explanations

web user agents

np_max-act · gemini-2.0-flash

This neuron activates on numeric tokens—especially version numbers and other decimal or digit-containing strings.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nesia

-0.07

 вра

-0.07

.RowHeaders

-0.07

Cab

-0.07

 rift

-0.07

 vice

-0.07

ERAL

-0.07

 eight

-0.06

NT

-0.06

 Leah

-0.06

POSITIVE LOGITS

bí

0.07

кому

0.07

++++++++

0.06

isches

0.06

_SWITCH

0.06

pository

0.06

 _↵

0.06

cui

0.06

 dây

0.06

 midi

0.05

Activations Density 0.002%

web user agents

This neuron activates on numeric tokens—especially version numbers and other decimal or digit-containing strings.

No Comments

No Known Activations