INDEX

Explanations

time-related expressions that indicate a sense of continuity or change

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

time since or after

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

 ſch

-0.45

 itſelf

-0.45

 faſt

-0.44

 leſs

-0.42

 Bohemia

-0.42

ByDate

-0.41

Legende

-0.41

 aveug

-0.40

 stadt

-0.40

 ſmall

-0.40

POSITIVE LOGITS

wußt

0.50

自从

0.50

 since

0.49

fjspx

0.49

findpost

0.47

 sejak

0.45

since

0.45

 setelah

0.45

abstractmethod

0.45

enumi

0.44

Activations Density 0.022%

time-related expressions that indicate a sense of continuity or change

time since or after

No Comments

No Known Activations