INDEX

Explanations

phrases that indicate temporal relationships or moments when events occur, particularly concerning wrong timing or important circumstances

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

"* time" or "at * time" phrases

np_acts-logits-general · gemini-2.0-flash

at time

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

AsUp

-0.69

 newOwner

-0.69

Filmografie

-0.62

NameInMap

-0.61

 خارجية

-0.60

róleo

-0.58

AspNetCore

-0.58

PIB

-0.54

requireNonNull

-0.53

POSITIVE LOGITS

 time

1.24

 times

1.13

 moment

1.13

 momento

1.06

 Zeitpunkt

1.03

moment

0.97

time

0.93

 momentos

0.91

 timing

0.89

 момент

0.88

Activations Density 0.231%

phrases that indicate temporal relationships or moments when events occur, particularly concerning wrong timing or important circumstances

"* time" or "at * time" phrases

at time

No Comments

No Known Activations