INDEX

Explanations

Java code

oai_token-act-pair · gemini-2.0-flash

period

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.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

CodeAttribute

-0.65

 kaarangay

-0.62

tizado

-0.54

 €/

-0.54

 africaine

-0.53

 Toibin

-0.53

iolis

-0.51

EACH

-0.50

󠁣

-0.50

pecting

-0.49

POSITIVE LOGITS

PerformLayout

0.59

 بيها

0.54

 подо

0.45

javas

0.44

 Cros

0.43

Aiheesta

0.42

Bonnie

0.42

 endfor

0.42

PhysRevLett

0.42

Esc

0.41

Activations Density 0.138%

Java code

period

No Comments

No Known Activations