INDEX

Explanations

references to geographic locations or spatial descriptions

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

north/south/west side of

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

 الرياضيه

-0.49

setVerticalGroup

-0.46

 Wikimedijinoj

-0.41

FieldDescriptor

-0.41

 CURIAM

-0.41

AnchorStyles

-0.40

FXML

-0.40

 Autorisations

-0.39

DECREF

-0.38

 Signalez

-0.38

POSITIVE LOGITS

styleType

0.50

crossing

0.41

Tn

0.40

TokenNameLPAREN

0.40

ohist

0.40

ront

0.40

romyal

0.39

wl

0.39

 circons

0.38

manic

0.38

Activations Density 0.018%

references to geographic locations or spatial descriptions

north/south/west side of

No Comments

No Known Activations