INDEX

Explanations

specific relational phrases that indicate connections or dependencies between entities

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

possessive pronouns

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 незавершена

-0.72

GEBURTSDATUM

-0.65

 disambiguazione

-0.65

 mijne

-0.63

 nôtre

-0.63

AndEndTag

-0.59

rungsseite

-0.57

Personensuche

-0.56

SequentialGroup

-0.56

UnsafeEnabled

-0.55

POSITIVE LOGITS

his

1.09

 его

0.94

 jego

0.92

 그의

0.92

 their

0.86

 jeho

0.85

 його

0.82

sua

0.82

his

0.80

他的

0.79

Activations Density 0.224%

specific relational phrases that indicate connections or dependencies between entities

possessive pronouns

No Comments

No Known Activations