INDEX

Explanations

=device

np_max-act · gemini-2.0-flash

This neuron detects HTML document “head” boilerplate, in particular `<meta>` tags (especially the viewport meta) and related head‐section elements.

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

Iso

-0.06

bst

-0.06

�

-0.06

 hart

-0.06

="{{$

-0.06

itr

-0.06

fst

-0.06

 HttpStatus

-0.06

 deadliest

-0.06

 Unters

-0.06

POSITIVE LOGITS

 solutions

0.07

 sensation

0.07

 capit

0.07

-made

0.07

 Assuming

0.07

_EMAIL

0.07

_point

0.07

.Exists

0.07

 solve

0.06

 Email

0.06

Activations Density 0.008%

=device

This neuron detects HTML document “head” boilerplate, in particular `<meta>` tags (especially the viewport meta) and related head‐section elements.

No Comments

No Known Activations

=device

This neuron detects HTML document “head” boilerplate, in particular `<meta>` tags (especially the viewport meta) and related head‐section elements.

No Comments

No Known Activations