INDEX

Explanations

references to specific locations and institutions

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

Location names

np_acts-logits-general · gemini-2.0-flash

cities and states

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ब्रेकडाउन

-0.73

Personendaten

-0.71

 informée

-0.69

oa̍t

-0.69

CloseOperation

-0.68

دانشنامهٔ

-0.68

 ujednoznacz

-0.65

sizeCache

-0.64

 Мексичка

-0.63

 utafitiHapana

-0.62

POSITIVE LOGITS

 same

0.41

same

0.35

 selben

0.34

 Same

0.34

 же

0.33

 Versorgung

0.32

endida

0.32

Same

0.32

 Дата

0.31

 mismas

0.30

Activations Density 0.224%

references to specific locations and institutions

Location names

cities and states

No Comments

No Known Activations