INDEX

Explanations

quantitative measurements and comparisons in the text

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

Numbers followed/preceded by math symbols

np_acts-logits-general · gemini-2.0-flash

hexadecimal numbers and code

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

complexContent

-0.62

Xna

-0.55

reras

-0.52

itol

-0.52

 labs

-0.47

 poêle

-0.45

poi

-0.45

displayquote

-0.45

 Igreja

-0.44

 Auss

-0.43

POSITIVE LOGITS

tomation

0.68

قایناقلار

0.57

ۜ

0.56

Vidite

0.54

migrationBuilder

0.54

﴿

0.52

зулта

0.52

idän

0.52

 mobilisation

0.51

bewerken

0.51

Activations Density 0.263%

quantitative measurements and comparisons in the text

Numbers followed/preceded by math symbols

hexadecimal numbers and code

No Comments

No Known Activations