INDEX

Explanations

references to conflict and interrogation situations

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

groups of people

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

 longitudinal

-0.42

 bowiem

-0.40

 jednak

-0.38

ChildScrollView

-0.38

 servici

-0.37

ंदीखरीदारी

-0.37

 nevertheless

-0.36

iasaan

-0.36

 Conversely

-0.36

GEBURTSDATUM

-0.36

POSITIVE LOGITS

 للاسماء

0.60

richTextPanel

0.59

 المعيارى

0.59

adaptiveStyles

0.54

 ujednoznacz

0.53

MigrationBuilder

0.48

 intStringLen

0.48

 SwitchCompat

0.47

DOCTYPE

0.47

IntoConstraints

0.46

Activations Density 0.084%

references to conflict and interrogation situations

groups of people

No Comments

No Known Activations