INDEX

Explanations

specific formatting sequences used in structured document layout, likely related to tables or matrix representations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Code or formatting markup

np_acts-logits-general · gemini-2.0-flash

formatting and tabular data

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

تقاوى

-0.68

><!--

-0.61

`<

-0.59

PreferredItem

-0.56

tek

-0.56

-0.53

 произ

-0.53

}}}{\

-0.53

))^

-0.53

]]=

-0.52

POSITIVE LOGITS

 للاسماء

0.79

ंदीखरीदारी

0.69

ThemeProvider

0.66

NUMX

0.64

ientí

0.63

 विश्वसनीयता

0.62

^(@)

0.61

〢

0.61

 itſelf

0.60

yntaxException

0.60

Activations Density 0.074%

specific formatting sequences used in structured document layout, likely related to tables or matrix representations

Code or formatting markup

formatting and tabular data

No Comments

No Known Activations