INDEX

Explanations

phrases indicating a request or expression of gratitude

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

what about

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

 EconPapers

-0.46

 agencias

-0.46

 للمعارف

-0.46

 asociaciones

-0.43

 препратки

-0.43

 trasform

-0.39

rrggbb

-0.38

contentLoaded

-0.38

 linee

-0.37

 vermogen

-0.37

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.57

 defStyle

0.54

 CURLOPT

0.43

GEBURTS

0.42

WriteTagHelper

0.41

 ***!

0.41

ftagPool

0.41

thums

0.40

XmlIgnore

0.40

 incomplète

0.40

Activations Density 0.022%

phrases indicating a request or expression of gratitude

what about

No Comments

No Known Activations