INDEX

Explanations

references to the "ELF" method and related descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Akira

-0.75

ReusableCell

-0.75

 Agassi

-0.75

ecore

-0.74

NetworkInfo

-0.73

 Roswell

-0.71

 Pavia

-0.71

таратура

-0.71

Fie

-0.71

 Agamemnon

-0.71

POSITIVE LOGITS

 Fork

0.76

 UIKit

0.71

Ske

0.68

Ske

0.68

 Mach

0.67

Wales

0.64

Alabama

0.64

Gilles

0.64

Ker

0.63

Pem

0.63

Activations Density 0.791%

references to the "ELF" method and related descriptors

No Comments

No Known Activations