INDEX

Explanations

words that indicate actions or processes involving investigation or reconsideration

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

spatial prepositions and sequences

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

"":

-0.35

ˌ

-0.33

 vuol

-0.30

`<

-0.30

نمای

-0.30

 شورای

-0.30

出现在

-0.30

変な

-0.29

writerow

-0.29

//[

-0.28

POSITIVE LOGITS

 autorytatywna

0.83

 ProtoMessage

0.77

 EconPapers

0.75

 disambiguazione

0.71

TagMode

0.70

 arşivlendi

0.69

uxxxx

0.68

Personensuche

0.66

enumii

0.65

 Geſch

0.65

Activations Density 0.280%

words that indicate actions or processes involving investigation or reconsideration

spatial prepositions and sequences

No Comments

No Known Activations