INDEX

Explanations

phrases related to measurements and statistical data

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

Prepositions and numbers

np_acts-logits-general · gemini-2.0-flash

collection, place, history

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

unique

-0.42

 unique

-0.34

)‏

-0.32

 UNIQUE

-0.32

tonode

-0.32

ことができる

-0.30

*/),

-0.29

Delight

-0.29

 Hidro

-0.29

sorted

-0.29

POSITIVE LOGITS

rungsseite

0.81

 propOrder

0.68

 مشين

0.65

Datuak

0.62

AddTagHelper

0.59

المكان

0.58

 بيها

0.57

Története

0.55

Autoritní

0.53

uxxxx

0.53

Activations Density 0.871%

phrases related to measurements and statistical data

Prepositions and numbers

collection, place, history

No Comments

No Known Activations