INDEX

Explanations

references to names of illnesses or conditions

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

Capitalized word fragments

np_acts-logits-general · gemini-2.0-flash

proper nouns and technical terms

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

the

-0.59

mässig

-0.53

-0.52

 ejus

-0.50

in

-0.50

and

-0.50

off

-0.48

one

-0.48

Literatuur

-0.48

two

-0.47

POSITIVE LOGITS

RenderAtEndOf

1.28

Personensuche

0.96

LookAnd

0.91

########.

0.90

RegressionTest

0.90

InvalidProtocol

0.89

MigrationBuilder

0.85

 GenerationType

0.84

AnchorTagHelper

0.83

 समीक्षक

0.82

Activations Density 1.010%

references to names of illnesses or conditions

Capitalized word fragments

proper nouns and technical terms

No Comments

No Known Activations