INDEX

Explanations

mathematical expressions and numeric data

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

rejected classifications

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

 kasarigan

-0.59

jsxFileName

-0.58

DockStyle

-0.55

didSet

-0.51

httphttps

-0.47

 Autorizaciones

-0.47

SequentialGroup

-0.46

SourceChecksum

-0.46

RenderAtEndOf

-0.45

complexContent

-0.45

POSITIVE LOGITS

 myſelf

0.41

César

0.40

andowski

0.40

 CHtml

0.40

uktur

0.39

Havolalar

0.38

 ſeveral

0.38

copi

0.38

 débris

0.38

 Geist

0.38

Activations Density 0.559%

mathematical expressions and numeric data

rejected classifications

No Comments

No Known Activations