INDEX

Explanations

punctuation

np_max-act · gemini-2.0-flash

The neuron consistently lights up on numerical tokens and list‐style markers (digits, decimals, module numbers, “–” separators, durations, etc.), i.e. structured numeric or list identifiers.

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

-media

-0.07

ne

-0.07

appa

-0.06

-K

-0.06

 tạo

-0.06

ポ

-0.06

ροφορίες

-0.06

ими

-0.06

 Utility

-0.06

-tracking

-0.06

POSITIVE LOGITS

 Vulkan

0.06

divide

0.06

stdbool

0.06

",↵↵

0.06

 prejud

0.06

ragon

0.06

 význam

0.06

HAND

0.06

@d

0.06

 obsess

0.06

Activations Density 0.027%

punctuation

The neuron consistently lights up on numerical tokens and list‐style markers (digits, decimals, module numbers, “–” separators, durations, etc.), i.e. structured numeric or list identifiers.

No Comments

No Known Activations