INDEX

Explanations

symbols or placeholders used to represent non-standard or special content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

]")]

-0.73

 FetchType

-0.72

 насељу

-0.71

 NgModule

-0.70

Personensuche

-0.67

getClassLoader

-0.66

ManagedBean

-0.66

NUMX

-0.64

esModule

-0.62

UIControlState

-0.59

POSITIVE LOGITS

SequentialGroup

0.60

WireFormat

0.55

 дописавши

0.52

 Comprometido

0.49

 fractionation

0.48

esteld

0.47

TintMode

0.47

 insuffisamment

0.47

 Merz

0.47

 YOUNG

0.47

Activations Density 0.429%

symbols or placeholders used to represent non-standard or special content

No Comments

No Known Activations