INDEX

Explanations

references to bans and suspensions, particularly in a legal or disciplinary context

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

being removed or banned

np_acts-logits-general · gemini-2.0-flash

ejection and suspension

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 فريبيس

-0.66

 Exactos

-0.55

Personendaten

-0.55

 betweenstory

-0.55

setVerticalGroup

-0.54

 Obrador

-0.54

SourceChecksum

-0.53

WebElementEntity

-0.53

səhifə

-0.52

TagMode

-0.49

POSITIVE LOGITS

 expulsion

1.40

 suspension

1.37

 expelled

1.28

 Suspension

1.19

 suspended

1.18

 suspensions

1.14

suspension

1.13

Suspension

1.09

 expel

1.07

 SUSP

1.07

Activations Density 0.500%

references to bans and suspensions, particularly in a legal or disciplinary context

being removed or banned

ejection and suspension

No Comments

No Known Activations