INDEX

Explanations

references to locations and names in a context of various categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 referenties

-0.58

 "..\..\..\

-0.57

 calculées

-0.54

ABASE

-0.53

diman

-0.52

Thêm

-0.51

FailureListener

-0.51

smtp

-0.50

quelle

-0.50

iedler

-0.49

POSITIVE LOGITS

EndInit

0.74

CrossRef

0.71

BeginInit

0.70

зулта

0.70

:✨

0.66

GEBURTSDATUM

0.65

migrationBuilder

0.62

 cellpadding

0.62

})()

0.62

حوالہ

0.58

Activations Density 1.517%

references to locations and names in a context of various categories

No Comments

No Known Activations