INDEX

Explanations

references to observational activities, particularly watching and listening

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 retours

-0.54

 profondità

-0.50

neri

-0.50

PackageManager

-0.49

 felicità

-0.49

ñadir

-0.48

 tebal

-0.48

 InputDecoration

-0.48

 idéia

-0.48

нуться

-0.48

POSITIVE LOGITS

 watching

1.66

 watch

1.62

 watched

1.60

 Watched

1.56

Watching

1.54

watching

1.51

 Watching

1.49

 watches

1.46

Watched

1.42

 WATCH

1.42

Activations Density 0.150%

references to observational activities, particularly watching and listening

No Comments

No Known Activations