INDEX

Explanations

punctuation

np_max-act · gemini-2.0-flash

The neuron activates on numeric tokens—especially years, dates, counts, and other numbers—in prose.

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

 playback

-0.07

 것을

-0.06

 bundled

-0.06

_encoding

-0.06

́

-0.06

CREASE

-0.06

 maintained

-0.06

tep

-0.06

renc

-0.06

 franchise

-0.06

POSITIVE LOGITS

öffent

0.07

 gặp

0.06

(route

0.06

Som

0.06

 BITTE

0.06

्ब

0.06

 beneficiaries

0.06

 accom

0.06

GCC

0.06

riters

0.06

Activations Density 0.229%

punctuation

The neuron activates on numeric tokens—especially years, dates, counts, and other numbers—in prose.

No Comments

No Known Activations