INDEX

Explanations

temporal references related to dates and past events

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

dates and times

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

 الحره

-0.36

 Träger

-0.33

can

-0.31

IsMutable

-0.28

Бележки

-0.28

 occhi

-0.28

eye

-0.27

Спољашње

-0.27

 eyes

-0.27

 layers

-0.26

POSITIVE LOGITS

 betweenstory

0.60

 Weekly

0.56

 Houſe

0.55

February

0.54

verwijspagina

0.54

Weekly

0.53

intios

0.53

abstractmethod

0.52

 fortnight

0.52

ſelves

0.52

Activations Density 0.004%

temporal references related to dates and past events

dates and times

No Comments

No Known Activations