INDEX

Explanations

phrases indicating definition or explanation

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

what words mean

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 calendriers

-0.55

 ✭✭

-0.54

сылкі

-0.49

addContainerGap

-0.48

Personensuche

-0.47

RTLR

-0.46

PullParser

-0.45

BibitemShut

-0.44

 clim

-0.44

)•

-0.43

POSITIVE LOGITS

 means

0.92

means

0.87

 meaning

0.82

Means

0.79

 Means

0.75

 mean

0.75

 significa

0.74

 MEANS

0.73

 bedeutet

0.68

 berarti

0.68

Activations Density 0.062%

phrases indicating definition or explanation

what words mean

No Comments

No Known Activations