INDEX

Explanations

phrases that specify constraints or particular characteristics related to objects or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 originali

-0.59

 rootReducer

-0.59

 liefer

-0.59

 ostavi

-0.56

lastIndex

-0.55

 EconPapers

-0.55

 larmes

-0.54

ﾟ)

-0.53

✨:

-0.53

 déclar

-0.52

POSITIVE LOGITS

 certain

1.18

 Certain

1.08

 particular

1.06

 bestimmte

1.03

certain

1.01

Certain

1.00

 tertentu

1.00

 specific

0.97

particular

0.95

 bestimmten

0.94

Activations Density 0.430%

phrases that specify constraints or particular characteristics related to objects or situations

No Comments

No Known Activations