INDEX

Explanations

didn

np_max-act · gemini-2.0-flash

The neuron fires on evaluative or discourse-linking words and phrases that introduce judgments or explanations (e.g. “remains,” “due (to),” “thus,” “kind (of),” “runs,” etc.).

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

Recording

-0.07

uars

-0.06

 Anxiety

-0.06

nos

-0.06

ure

-0.06

 injected

-0.06

Zoo

-0.06

 BOOK

-0.06

ON

-0.06

POSITIVE LOGITS

Ў

0.06

€↵

0.06

StatusBar

0.06

 있음

0.06

있는

0.06

 scripted

0.06

 charismatic

0.06

 withString

0.06

 \↵↵

0.06

apsed

0.06

Activations Density 0.024%

didn

The neuron fires on evaluative or discourse-linking words and phrases that introduce judgments or explanations (e.g. “remains,” “due (to),” “thus,” “kind (of),” “runs,” etc.).

No Comments

No Known Activations