INDEX

Explanations

before/will

np_max-act · gemini-2.0-flash

The neuron fires on tokens that mark “before” or impending‐action event/callback names (e.g. beforeXxx, onBeforeXxx, will_change).

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

.controls

-0.07

 muchas

-0.07

 Publish

-0.07

_CHANNEL

-0.07

-0.06

(key

-0.06

 diarr

-0.06

_RADIO

-0.06

 study

-0.06

_COLLECTION

-0.06

POSITIVE LOGITS

swagen

0.07

带

0.07

Pro

0.07

ropping

0.06

distinct

0.06

婚

0.06

-net

0.06

髪

0.06

ри

0.06

 правильно

0.06

Activations Density 0.009%

before/will

The neuron fires on tokens that mark “before” or impending‐action event/callback names (e.g. beforeXxx, onBeforeXxx, will_change).

No Comments

No Known Activations