INDEX

Explanations

structural elements or markers in a document, possibly signaling the structure of a code or formatted data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 betweenstory

-0.85

 ujednoznacz

-0.69

buttonBar

-0.69

 linkovi

-0.68

లాలు

-0.68

 ProtoMessage

-0.67

AndEndTag

-0.67

 '\\;'

-0.67

RegressionTest

-0.66

urlpatterns

-0.66

POSITIVE LOGITS

NOPQRST

0.59

 Erwartungen

0.52

 CreateTagHelper

0.50

 pinulongan

0.49

!*\

0.48

ebenarnya

0.48

 paroi

0.48

Faktor

0.48

MessageTagHelper

0.46

près

0.46

Activations Density 0.026%

structural elements or markers in a document, possibly signaling the structure of a code or formatted data

No Comments

No Known Activations