INDEX

Explanations

comparisons

np_max-act · gemini-2.0-flash

This neuron activates on comparative “similar to” constructions (i.e. when the text is drawing an explicit analogy).

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

 Harold

-0.08

ائف

-0.07

idon

-0.07

 Quickly

-0.07

respect

-0.07

McG

-0.07

 çağ

-0.06

estar

-0.06

approx

-0.06

 swiftly

-0.06

POSITIVE LOGITS

<=$

0.07

pn

0.06

 jogador

0.06

inois

0.06

 подключ

0.06

 Cord

0.06

 eerie

0.06

ondere

0.06

 своє

0.06

_IF

0.06

Activations Density 0.047%

comparisons

This neuron activates on comparative “similar to” constructions (i.e. when the text is drawing an explicit analogy).

No Comments

No Known Activations