INDEX

Explanations

you

np_max-act · gemini-2.0-flash

The neuron strongly activates on expressions of thanks or gratitude (e.g., “thank you,” “thanks”), flagging places where reviewers or writers are offering appreciation.

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

[K

-0.07

specified

-0.06

KY

-0.06

Ky

-0.06

 kalk

-0.06

mann

-0.06

δοση

-0.06

經

-0.06

 Biol

-0.06

Attrib

-0.06

POSITIVE LOGITS

.INSTANCE

0.08

 Outer

0.07

Fer

0.06

ocyte

0.06

.ENTER

0.06

Enterprise

0.06

 Tables

0.06

_inner

0.06

 sturdy

0.06

inactive

0.06

Activations Density 0.062%

you

The neuron strongly activates on expressions of thanks or gratitude (e.g., “thank you,” “thanks”), flagging places where reviewers or writers are offering appreciation.

No Comments

No Known Activations