INDEX

Explanations

indicators of significant events or changes

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

journalistic sources

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

Reprodução

-0.57

 mentiras

-0.57

createCanvas

-0.56

 defaultstate

-0.56

queles

-0.54

 henkil

-0.53

 للمعارف

-0.52

 esclavos

-0.51

 ciudadanía

-0.51

 generaciones

-0.51

POSITIVE LOGITS

 Reuters

0.45

 Indic

0.43

bnf

0.43

 Joint

0.42

 indicated

0.41

afp

0.38

GOTREF

0.38

joint

0.37

 espect

0.37

 filo

0.37

Activations Density 0.001%

indicators of significant events or changes

journalistic sources

No Comments

No Known Activations