INDEX

Explanations

to

np_max-act · gemini-2.0-flash

This neuron detects words expressing ongoing or persistent states, particularly the phrase “continues to be.”

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

-mediated

-0.07

Probability

-0.07

 également

-0.07

ynamic

-0.06

Level

-0.06

ête

-0.06

 mediated

-0.06

Coefficient

-0.06

 exempl

-0.06

olidays

-0.06

POSITIVE LOGITS

шее

0.07

 Hamas

0.06

 bitte

0.06

 затвердж

0.06

_EQUAL

0.06

�

0.06

 України

0.06

(stock

0.06

训

0.06

雙

0.06

Activations Density 0.056%

to

This neuron detects words expressing ongoing or persistent states, particularly the phrase “continues to be.”

No Comments

No Known Activations