INDEX

Explanations

symbols or characters used in programming syntax

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

categories, fields, and params

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 avoient

-0.56

 pouvoit

-0.43

 auroit

-0.42

 başladı

-0.41

 sauter

-0.41

 feroit

-0.39

InstrumentedTest

-0.39

launch

-0.39

 économ

-0.38

 étoient

-0.37

POSITIVE LOGITS

mappedBy

0.58

ruptedException

0.57

 RoutedEventArgs

0.56

béco

0.56

Hochspringen

0.53

 EventArgs

0.53

 ویکی‌پدیای

0.51

tamol

0.51

μβρίου

0.50

atimes

0.48

Activations Density 0.007%

symbols or characters used in programming syntax

categories, fields, and params

No Comments

No Known Activations