INDEX

Explanations

words related to quantities, helpfulness, or attraction

oai_token-act-pair · gemini-2.0-flash

Code and filenames

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Anſ

-0.60

umpang

-0.56

 فريبيس

-0.56

 ―――――

-0.56

 myſelf

-0.54

 lèvres

-0.54

 itſelf

-0.54

 Ingo

-0.53

 vérit

-0.53

BibitemShut

-0.53

POSITIVE LOGITS

 "..\..\..\

0.57

astéroïdes

0.55

 OFDb

0.54

!*\

0.53

 برانيه

0.53

>//

0.51

 "..\..\

0.51

irited

0.50

SpringRunner

0.49

pushFollow

0.48

Activations Density 8.783%

words related to quantities, helpfulness, or attraction

Code and filenames

No Comments

No Known Activations