INDEX

Explanations

phrases related to experimental or research discussions

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

for instance

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

 للمعارف

-0.70

出版年

-0.59

AMILY

-0.59

 nahilalakip

-0.56

fieras

-0.54

 nonUne

-0.53

IsContent

-0.51

domada

-0.50

Tikang

-0.50

 Houſe

-0.49

POSITIVE LOGITS

*/;

0.35

PropertyChanged

0.32

findpost

0.32

 instinctive

0.31

HtmlAttribute

0.31

日閲覧

0.30

Erreferentziak

0.29

XmlAccessType

0.29

RefNanny

0.29

InjectMocks

0.29

Activations Density 0.014%

phrases related to experimental or research discussions

for instance

No Comments

No Known Activations

phrases related to experimental or research discussions

for instance

No Comments

No Known Activations