INDEX

Explanations

connections and relationships between concepts or entities in a context

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

resulted in, contained some, depend linearly

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

 للمعارف

-0.54

 saites

-0.46

حياته

-0.46

MVH

-0.45

 صوتيه

-0.44

hyrchwyd

-0.44

 detailsLots

-0.43

Życiorys

-0.42

GeoNames

-0.41

 للاسماء

-0.41

POSITIVE LOGITS

InjectAttribute

0.57

featureID

0.55

tvguidetime

0.44

StoryboardSegue

0.43

forChild

0.42

 ExecuteAsync

0.41

หมาย

0.40

VersionUID

0.40

migrationBuilder

0.40

 CommonModule

0.39

Activations Density 0.163%

connections and relationships between concepts or entities in a context

resulted in, contained some, depend linearly

No Comments

No Known Activations