INDEX

Explanations

phrases indicating possession or capability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

-0.84

Tembelea

-0.77

Efq

-0.77

reportWebVitals

-0.73

 houſe

-0.72

RectangleBorder

-0.71

 thither

-0.68

 ſeveral

-0.65

 Houſe

-0.65

 becauſe

-0.64

POSITIVE LOGITS

mtd

0.67

 advantage

0.67

 الوس

0.62

/*

0.60

 same

0.58

 Jacobsen

0.55

 gevol

0.55

//

0.55

</strong>

0.55

 necessary

0.54

Activations Density 0.220%

phrases indicating possession or capability

No Comments

No Known Activations