INDEX

Explanations

URLs and web addresses

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

URLs with paths and extensions

np_acts-logits-general · gemini-2.0-flash

URLs and academic citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 auffi

-0.69

piu

-0.60

SuppressMessage

-0.60

 deſt

-0.59

 purpoſe

-0.58

ValueStyle

-0.58

 simila

-0.58

 similarly

-0.57

 juſ

-0.56

 cultur

-0.55

POSITIVE LOGITS

 مشين

0.63

AxisAlignment

0.63

oriam

0.61

 EconPapers

0.60

’.”

0.59

).”

0.58

出版年

0.57

!—

0.57

Activations Density 0.081%

URLs and web addresses

URLs with paths and extensions

URLs and academic citations

No Comments

No Known Activations