INDEX

Explanations

with com

np_max-act · gemini-2.0-flash

The neuron activates on the special instruction markers (“||” and “>>”) that indicate lines to be left in English.

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

dd

-0.06

_notification

-0.06

ิพ

-0.06

 Alert

-0.06

.qty

-0.06

غن

-0.06

.Reverse

-0.06

"{{

-0.06

ищ

-0.06

واره

-0.06

POSITIVE LOGITS

 Liter

0.07

_def

0.06

TokenType

0.06

OMG

0.06

_RESET

0.06

Mother

0.06

took

0.06

 minist

0.06

-offsetof

0.06

 جن

0.06

Activations Density 0.012%

with com

The neuron activates on the special instruction markers (“||” and “>>”) that indicate lines to be left in English.

No Comments

No Known Activations