INDEX

Explanations

phrases related to updates and accuracy in information

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

"Up to date" or "up-to-date" phrases

np_acts-logits-general · gemini-2.0-flash

up to date

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 oprot

-0.59

__":

-0.57

 kasarigan

-0.55

momix

-0.55

 appContext

-0.54

phia

-0.54

nitř

-0.53

 مشارکت‌کنندگان

-0.51

 חיצוניים

-0.51

__":

-0.51

POSITIVE LOGITS

 current

1.03

 updated

0.94

current

0.92

 Current

0.92

 dated

0.87

Current

0.87

 date

0.87

 upto

0.85

 CURRENT

0.83

date

0.83

Activations Density 0.158%

phrases related to updates and accuracy in information

"Up to date" or "up-to-date" phrases

up to date

No Comments

No Known Activations