INDEX

Explanations

web content

np_max-act · gemini-2.0-flash

This neuron detects placeholder tokens (like “NAME_” and associated digits) used for slot-filling or template fields.

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

_coords

-0.07

 Mish

-0.07

 shirts

-0.07

 Eigen

-0.06

imeInterval

-0.06

-ing

-0.06

_pos

-0.06

یز

-0.06

_bold

-0.06

اضی

-0.06

POSITIVE LOGITS

 uneven

0.07

 машин

0.06

 Molly

0.06

 antagon

0.06

 royalty

0.06

 Worker

0.06

 tyranny

0.06

lei

0.06

ूछ

0.06

ISTRATION

0.06

Activations Density 0.023%

web content

This neuron detects placeholder tokens (like “NAME_” and associated digits) used for slot-filling or template fields.

No Comments

No Known Activations