INDEX

Explanations

names and titles of individuals or entities

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

Cattaneo French response

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

gonic

-0.63

*>(

-0.58

sizeCache

-0.58

 BrowserModule

-0.57

:\/\/

-0.57

ukunft

-0.56

esModule

-0.56

Orient

-0.56

horizon

-0.55

omite

-0.55

POSITIVE LOGITS

pp

0.56

ett

0.52

ETT

0.51

 Bigg

0.51

ll

0.50

 nett

0.50

 Técnicas

0.50

MMdd

0.49

Abb

0.48

llll

0.48

Activations Density 0.311%

names and titles of individuals or entities

Cattaneo French response

No Comments

No Known Activations