INDEX

Explanations

hexadecimal representations of code or data elements

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

WebElementEntity

-0.71

 विश्वसनीयता

-0.60

 Waray

-0.58

IUrlHelper

-0.58

angue

-0.52

 betweenstory

-0.51

 rather

-0.51

casio

-0.50

鲜

-0.49

EndContext

-0.49

POSITIVE LOGITS

xe

1.15

xc

1.14

xb

1.09

xC

1.02

xf

1.02

xd

1.01

xB

0.99

xa

0.99

xD

0.97

xE

0.96

Activations Density 0.140%

hexadecimal representations of code or data elements

No Comments

No Known Activations