INDEX

Explanations

parts of code and instructions, occasionally activating on related words such as "tips" or "models"

oai_token-act-pair · gemini-2.0-flash

string

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ConverterFactory

-0.52

ınıza

-0.44

曖昧さ回避

-0.41

디오

-0.40

ளை

-0.40

insertBefore

-0.39

러나

-0.39

vejte

-0.39

astéroïdes

-0.39

ุทธ

-0.38

POSITIVE LOGITS

<bos>

1.05

IsContent

1.00

AddTagHelper

0.93

 تضيفلها

0.74

TagMode

0.72

 виправивши

0.72

bootstrapcdn

0.71

 تانيه

0.69

 AssemblyCulture

0.69

DoubleQuotes

0.68

Activations Density 1.214%

parts of code and instructions, occasionally activating on related words such as "tips" or "models"

string

No Comments

No Known Activations