INDEX

Explanations

songs and recordings

np_max-act · gemini-2.0-flash

The neuron predominantly activates on numeric expressions (especially years or other numbers).

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

crit

-0.08

pdf

-0.07

 ErrorCode

-0.07

Theme

-0.06

charges

-0.06

_attention

-0.06

 restart

-0.06

 Fault

-0.06

ground

-0.06

ми

-0.06

POSITIVE LOGITS

Ascii

0.07

 مبار

0.07

;');↵

0.07

Updater

0.06

eldon

0.06

 비교

0.06

 sentido

0.06

Roman

0.06

 möglich

0.06

quia

0.06

Activations Density 0.027%

songs and recordings

The neuron predominantly activates on numeric expressions (especially years or other numbers).

No Comments

No Known Activations