INDEX

Explanations

phrases that express possession and preference, often with emotional or physical emphasis

oai_token-act-pair · gemini-2.0-flash

words related to design

np_token-act-pair-logits · gpt-4o-mini

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 "..\..\..\

-0.97

uxxxx

-0.79

ruptedException

-0.72

 "..\..\

-0.71

 مشين

-0.70

aarrggbb

-0.69

#+#

-0.68

__":

-0.68

脚注の使い方

-0.67

__':

-0.67

POSITIVE LOGITS

 mind

0.62

fit

0.56

 Designs

0.48

 designs

0.47

designs

0.45

เสีย

0.45

IER

0.43

chey

0.42

ose

0.42

ább

0.42

Activations Density 1.641%

phrases that express possession and preference, often with emotional or physical emphasis

words related to design

No Comments

No Known Activations