INDEX

Explanations

important

np_max-act · gemini-2.0-flash

This neuron activates on superlative or emphasis words signaling standout prominence (e.g. “most,” “prime,” “notable,” “widely-known”).

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

 Indonesian

-0.07

 mistaken

-0.06

��

-0.06

 Unsure

-0.06

 Indonesia

-0.06

Insensitive

-0.06

_dma

-0.06

(tol

-0.06

gua

-0.06

POSITIVE LOGITS

 rhythm

0.07

reck

0.07

docs

0.06

="{

0.06

 Initialize

0.06

 adventures

0.06

Sql

0.06

isOpen

0.06

 reducers

0.06

 prelim

0.06

Activations Density 0.081%

important

This neuron activates on superlative or emphasis words signaling standout prominence (e.g. “most,” “prime,” “notable,” “widely-known”).

No Comments

No Known Activations