INDEX

Explanations

affiliates

np_max-act · gemini-2.0-flash

The neuron fires on the license‐header token “affiliates,” i.e. it detects the occurrence of “affiliates” (as in “Oracle and/or its affiliates”) in file headers.

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

="#">↵

-0.07

 jedoch

-0.07

 наиболее

-0.06

 invert

-0.06

PLIED

-0.06

.runner

-0.06

ียนร

-0.06

 eget

-0.06

 assistir

-0.06

imeType

-0.06

POSITIVE LOGITS

 affiliates

0.07

 Orientation

0.06

 cowork

0.06

 concentrates

0.06

 число

0.06

 Mine

0.06

-force

0.06

حث

0.06

사무

0.06

,/

0.06

Activations Density 0.000%

affiliates

The neuron fires on the license‐header token “affiliates,” i.e. it detects the occurrence of “affiliates” (as in “Oracle and/or its affiliates”) in file headers.

No Comments

No Known Activations