INDEX

Explanations

It seems like this neuron is very complex, but the main thing it does is find bits of code from a variety of languages while giving more importance to parts of Java and C++

oai_token-act-pair · gemini-2.0-flash

parentheses/brackets

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

ArgsConstructor

-0.68

INCREF

-0.52

ซ์

-0.49

 photo

-0.47

otheses

-0.47

urp

-0.46

(@"%@",

-0.46

AndPassword

-0.46

appId

-0.45

tesis

-0.45

POSITIVE LOGITS

CppMethod

0.65

 InputDecoration

0.60

 avoient

0.59

 étoient

0.59

Personendaten

0.56

 étoit

0.56

oredCriteria

0.56

 wikipagina

0.55

iecie

0.55

TintMode

0.54

Activations Density 1.178%

It seems like this neuron is very complex, but the main thing it does is find bits of code from a variety of languages while giving more importance to parts of Java and C++

parentheses/brackets

No Comments

No Known Activations