INDEX

Explanations

references to specific components or systems, particularly labeled with a "Q" designation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Wicidata

-0.62

 transfieras

-0.56

principalColumn

-0.55

 CanadaChoose

-0.54

)}+\

-0.52

Diwedd

-0.52

клопе

-0.52

Autoritní

-0.52

 kohdetta

-0.52

Бахар

-0.51

POSITIVE LOGITS

3.38

3.16

1.96

Ｑ

1.77

Qs

1.60

1.43

Qn

1.25

𝑄

1.25

QH

1.21

QQ

1.18

Activations Density 0.138%

references to specific components or systems, particularly labeled with a "Q" designation

No Comments

No Known Activations

references to specific components or systems, particularly labeled with a "Q" designation

No Comments

No Known Activations