INDEX

Explanations

elements related to performance and visibility in different contexts

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

"On [something]" or references to stage or screens

np_acts-logits-general · gemini-2.0-flash

public performance or display

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

aarrggbb

-0.60

 Entra

-0.47

 deten

-0.47

umba

-0.46

arote

-0.42

 пет

-0.42

WriteAttribute

-0.42

Lowest

-0.41

 buried

-0.41

 CreateTagHelper

-0.40

POSITIVE LOGITS

 screen

1.40

 stage

1.32

 screens

1.27

screen

1.23

stage

1.13

creen

1.12

creens

1.10

 Screens

1.09

 Screen

1.09

 SCREEN

1.08

Activations Density 0.103%

elements related to performance and visibility in different contexts

"On [something]" or references to stage or screens

public performance or display

No Comments

No Known Activations