INDEX

Explanations

references to time-related events and conditions

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

after a condition changes

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

 doccia

-0.38

FTFY

-0.38

 Displays

-0.34

 camicia

-0.32

 Offisielt

-0.32

 pintados

-0.32

setDefault

-0.31

kembangan

-0.31

owskiego

-0.30

oron

-0.30

POSITIVE LOGITS

parsedMessage

0.63

WithIOException

0.57

 jspb

0.54

ItemBackground

0.50

IntoConstraints

0.50

WaitGroup

0.49

verifyException

0.49

getColumnIndex

0.48

UnsafeEnabled

0.47

 iprot

0.47

Activations Density 0.081%

references to time-related events and conditions

after a condition changes

No Comments

No Known Activations