INDEX

Explanations

references to fraud, theft, corruption, and criminal activity

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

accusations of fraud, stealing, abuse, corruption, theft

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

TRIBUN

-0.49

WritableDatabase

-0.48

 jsPsych

-0.43

ieteur

-0.40

twimg

-0.40

 Gefahr

-0.39

tagHelper

-0.38

ajara

-0.38

 flattered

-0.38

inSlope

-0.38

POSITIVE LOGITS

URLException

0.40

 Italijani

0.39

SBATCH

0.39

Referencie

0.37

 concerning

0.36

StackNavigator

0.35

🔀

0.35

寞

0.35

вгений

0.35

DNEY

0.35

Activations Density 0.185%

references to fraud, theft, corruption, and criminal activity

accusations of fraud, stealing, abuse, corruption, theft

No Comments

No Known Activations