INDEX

Explanations

information related to service agreements and customer compensation policies

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

"For" followed by positive words

np_acts-logits-general · gemini-2.0-flash

reward for effort or status

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 StatefulWidget

-0.82

UserScript

-0.66

 StatelessWidget

-0.66

]^{-

-0.62

 للمعارف

-0.59

ंदीखरीदारी

-0.59

öglichkeiten

-0.58

 Obrador

-0.57

ngdoc

-0.56

 possibili

-0.55

POSITIVE LOGITS

 successful

0.78

successful

0.68

Successful

0.64

 successfully

0.63

 qualifying

0.63

 proven

0.59

 réussite

0.58

 Successful

0.57

 aank

0.56

 terbukti

0.55

Activations Density 0.251%

information related to service agreements and customer compensation policies

"For" followed by positive words

reward for effort or status

No Comments

No Known Activations