INDEX

Explanations

specifications related to width in design or layout contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

HTML image width attribute

np_acts-logits-general · gemini-2.0-flash

width followed by = or :

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RTEX

-0.59

:\/\/

-0.58

 Gurney

-0.57

 iArr

-0.57

ThroughAttribute

-0.55

 resourceCulture

-0.55

 uVar

-0.55

Endnotes

-0.54

ette

-0.54

ziger

-0.54

POSITIVE LOGITS

 width

1.04

 widths

0.98

width

0.95

 Width

0.93

WIDTH

0.91

 WIDTH

0.90

drawal

0.87

Wid

0.85

Width

0.83

 maxWidth

0.78

Activations Density 0.049%

specifications related to width in design or layout contexts

HTML image width attribute

width followed by = or :

No Comments

No Known Activations