INDEX

Explanations

true

np_max-act · gemini-2.0-flash

The neuron flags code‐style identifiers—especially Android resource or API names containing dots (e.g. “R.id…”, “getMenuInflater()”, etc.).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

imes

-0.06

.experimental

-0.06

 Henri

-0.06

 reife

-0.06

 Irene

-0.06

руш

-0.06

PUT

-0.06

DataProvider

-0.06

ати

-0.06

중에

-0.06

POSITIVE LOGITS

 sailing

0.07

 none

0.07

 Serialize

0.07

	assertEquals

0.07

 temas

0.07

 zIndex

0.07

 calculation

0.06

 advertised

0.06

인은

0.06

式会社

0.06

Activations Density 0.003%

true

The neuron flags code‐style identifiers—especially Android resource or API names containing dots (e.g. “R.id…”, “getMenuInflater()”, etc.).

No Comments

No Known Activations