INDEX

Explanations

phrases related to accountability and observation in various contexts

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

dealing with technical or foreign terms

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

skab

-0.42

 sice

-0.42

<h1>

-0.39

UNDA

-0.39

ža

-0.38

 observational

-0.37

-0.36

leſs

-0.36

 serai

-0.36

ARTA

-0.35

POSITIVE LOGITS

SourceChecksum

0.70

RectangleBorder

0.62

 Taktlose

0.61

IUrlHelper

0.58

 مشين

0.57

évaluateur

0.56

DockStyle

0.55

WithIOException

0.54

 CreateTagHelper

0.53

GIVEREF

0.53

Activations Density 0.024%

phrases related to accountability and observation in various contexts

dealing with technical or foreign terms

No Comments

No Known Activations