INDEX

Explanations

terms related to time and measurements

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

temporal variation

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

ArgsConstructor

-0.41

bac

-0.38

DockStyle

-0.38

henswürdigkeiten

-0.36

HttpServlet

-0.35

 Affirm

-0.35

vertret

-0.35

mé

-0.34

 volk

-0.33

walde

-0.33

POSITIVE LOGITS

ConstraintMaker

0.64

RenderAtEndOf

0.63

MemoryWarning

0.59

writeFieldEnd

0.56

 OFDb

0.54

XtraBars

0.53

 resourceCulture

0.52

adaptiveStyles

0.50

 getRule

0.50

sizeCache

0.49

Activations Density 0.210%

terms related to time and measurements

temporal variation

No Comments

No Known Activations