INDEX

Explanations

one

np_max-act · gemini-2.0-flash

The neuron activates specifically on numeric tokens, i.e. standalone numbers appearing in the text.

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

儀

-0.08

Exp

-0.07

 salad

-0.07

rais

-0.06

_ts

-0.06

_z

-0.06

_va

-0.06

tar

-0.06

jpg

-0.06

Ro

-0.06

POSITIVE LOGITS

 urlencode

0.08

ERVE

0.07

 التش

0.07

 منتشر

0.07

handled

0.06

 ingress

0.06

.CurrentCulture

0.06

 chờ

0.06

 Incontri

0.06

_AUT

0.06

Activations Density 0.025%

one

The neuron activates specifically on numeric tokens, i.e. standalone numbers appearing in the text.

No Comments

No Known Activations