INDEX

Explanations

code blocks, likely related to conditional statements or loops

oai_token-act-pair · gemini-2.0-flash

Code or foreign language

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

Not in Any Lists

No Comments

Negative Logits

 itſelf

-0.78

queryInterface

-0.75

 pleaſure

-0.74

 myſelf

-0.73

ſelf

-0.73

ItemLayout

-0.71

SequentialGroup

-0.69

AndEndTag

-0.69

 Theſe

-0.69

 kasarigan

-0.67

POSITIVE LOGITS

necedor

0.59

 kirja

0.55

WebElementEntity

0.55

0.53

0.52

webElementXpaths

0.48

syms

0.47

<unused62>

0.46

اعدة

0.45

Activations Density 4.541%

code blocks, likely related to conditional statements or loops

Code or foreign language

No Comments

No Known Activations