INDEX

Explanations

commas

np_max-act · gemini-2.0-flash

The neuron activates on single‐word category or tag labels (e.g. “Green,” “BMW,” “Hatchback,” “Electric”) in metadata sections.

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

 entert

-0.07

 toolbar

-0.07

 decorations

-0.06

_PG

-0.06

 Madd

-0.06

PRETTY

-0.06

ิญ

-0.06

.animations

-0.06

 центра

-0.06

ti

-0.06

POSITIVE LOGITS

lid

0.07

'';

0.06

 이동합니다

0.06

.Embed

0.06

Documento

0.06

(pi

0.06

ятся

0.06

 Crypt

0.06

ordan

0.06

 calam

0.06

Activations Density 0.012%

commas

The neuron activates on single‐word category or tag labels (e.g. “Green,” “BMW,” “Hatchback,” “Electric”) in metadata sections.

No Comments

No Known Activations