INDEX

Explanations

references to certain companies or entities, particularly those starting with "Pers."

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

words beginning with Pers

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 itſelf

-0.54

 Ubicación

-0.47

nait

-0.45

IOException

-0.44

 Accommodations

-0.44

Nominations

-0.44

Enrollment

-0.43

galkan

-0.43

 idolat

-0.43

elakaan

-0.42

POSITIVE LOGITS

 Pers

2.50

Pers

2.39

pers

2.25

 pers

2.20

 PERS

1.96

PERS

1.56

 persu

1.16

 persuasion

1.12

 persisted

1.08

 Perse

1.00

Activations Density 0.005%

references to certain companies or entities, particularly those starting with "Pers."

words beginning with Pers

No Comments

No Known Activations