INDEX

Explanations

numerical data or references to specific values

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

numerical data or statistics within the text

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

numbers and abbreviations like CE and Inc

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndroidJUnit

-0.81

 INTERESAR

-0.65

Expedia

-0.61

rungsseite

-0.54

 combineReducers

-0.54

 Roskov

-0.52

UnknownFields

-0.52

 تضيفلها

-0.51

transQ

-0.51

 Xperia

-0.51

POSITIVE LOGITS

 ModelExpression

0.52

 sunrise

0.50

 Dage

0.46

UITableViewCell

0.46

AndEndTag

0.46

 rails

0.45

LogLevel

0.45

TypedDataSet

0.43

nę

0.43

rail

0.43

Activations Density 0.139%

numerical data or references to specific values

numerical data or statistics within the text

numbers and abbreviations like CE and Inc

No Comments

No Known Activations