INDEX

Explanations

the word "real" and superlatives

oai_token-act-pair · gemini-2.0-flash

real

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

AndEndTag

-1.02

 raiſ

-1.00

 Monfieur

-0.85

 resourceCulture

-0.85

 Houſe

-0.84

Personendaten

-0.84

 Anſ

-0.83

 صوتيه

-0.82

 becauſe

-0.81

 المعيارى

-0.81

POSITIVE LOGITS

rawDesc

0.46

bori

0.44

STATEMENT

0.42

welle

0.41

ワー

0.41

Act

0.40

 STATEMENT

0.40

araf

0.39

 anni

0.38

tableName

0.38

Activations Density 3.532%

the word "real" and superlatives

real

No Comments

No Known Activations