INDEX

Explanations

sections of code or programming syntax

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

end_of_turn / start_of_turn

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

XmlAccessorType

-0.58

 '\\;'

-0.57

principalColumn

-0.53

/*

-0.53

AddTagHelper

-0.53

EDEFAULT

-0.52

LookAnd

-0.51

'{@

-0.50

GEBURTSDATUM

-0.50

 AssemblyProduct

-0.49

POSITIVE LOGITS

MERCIAL

0.40

 Miranda

0.38

 Incomplete

0.37

bif

0.37

marginHorizontal

0.36

incomplete

0.36

ablemente

0.36

 Virat

0.35

 faulty

0.35

Incomplete

0.34

Activations Density 0.013%

sections of code or programming syntax

end_of_turn / start_of_turn

No Comments

No Known Activations