INDEX

Explanations

commas

np_max-act · gemini-2.0-flash

This neuron selectively activates on the special “<|start_header_id|>” tokens that mark the beginning of a message header.

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

.enterprise

-0.07

 rhyth

-0.07

Denied

-0.06

EK

-0.06

_allocator

-0.06

_ap

-0.06

 Leopard

-0.06

 pact

-0.06

िल

-0.06

フト

-0.06

POSITIVE LOGITS

accuracy

0.06

.Image

0.06

 світі

0.06

.visitMethodInsn

0.06

ERİ

0.06

.chdir

0.06

 streamline

0.06

 rugby

0.06

 tipping

0.06

.HtmlControls

0.06

Activations Density 0.012%

commas

This neuron selectively activates on the special “<|start_header_id|>” tokens that mark the beginning of a message header.

No Comments

No Known Activations

commas

This neuron selectively activates on the special “<|start_header_id|>” tokens that mark the beginning of a message header.

No Comments

No Known Activations