INDEX

Explanations

a combination of seemingly unrelated character sequences, possibly related to academic references, medical terminology, and numerical identifiers

oai_token-act-pair · gemini-2.0-flash

code

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

prepareStatement

-0.64

GraphicsUnit

-0.60

ResumeLayout

-0.60

createStatement

-0.59

phazard

-0.59

 NgModule

-0.56

 ligiloj

-0.54

tonode

-0.54

ollectionView

-0.54

ingeki

-0.54

POSITIVE LOGITS

FMS

0.57

xase

0.55

 Tully

0.52

 Dandy

0.50

 AssemblyTitle

0.50

AddHtmlAttribute

0.49

 старости

0.49

 pleins

0.49

xxl

0.49

balleur

0.49

Activations Density 0.961%

a combination of seemingly unrelated character sequences, possibly related to academic references, medical terminology, and numerical identifiers

code

No Known Activations