INDEX

Explanations

sequences of whitespace characters

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

XML and HTML tags

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



-0.59

ImageContext

-0.56

GraphicsUnit

-0.54

errHandler

-0.54

addGap

-0.52

ValueStyle

-0.52

 BoxFit

-0.50

+#+#

-0.50

 ThemeData

-0.49

StructEnd

-0.49

POSITIVE LOGITS

MigrationBuilder

0.53

complexContent

0.42

rida

0.42

utin

0.42

 Waray

0.41

dymyr

0.41

 resolutely

0.41

ṃ

0.40

 besonder

0.40

feitura

0.40

Activations Density 0.007%

sequences of whitespace characters

XML and HTML tags

No Comments

No Known Activations