INDEX

Explanations

temporal markers indicating recent events

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

recent past time references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 otomatig

-0.43

 ویکی‌آمباردا

-0.42

afficheront

-0.40

!*\

-0.40

 ویکی‌پدی

-0.38

Rè

-0.38

parsedMessage

-0.37

 odkazy

-0.36

verschluss

-0.36

tahankan

-0.36

POSITIVE LOGITS

IntoConstraints

0.63

ftagPool

0.43

 recently

0.42

 iprot

0.42

épaisseur

0.40

 Numerade

0.40

FieldNumber

0.40

になって

0.40

 pygame

0.39

 récemment

0.39

Activations Density 0.022%

temporal markers indicating recent events

recent past time references

No Comments

No Known Activations