INDEX

Explanations

instances of the word "walk" and its variations, indicating a focus on walking actions and activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 particular

-0.43

 potência

-0.41

endregion

-0.40

 Aguilera

-0.38

 secret

-0.37

 especiais

-0.36

 suppose

-0.36

subpackage

-0.36

 specific

-0.36

 certain

-0.35

POSITIVE LOGITS

Walk

1.38

 WALK

1.33

 Walk

1.32

walk

1.27

 walk

1.27

 Walks

1.27

 Walking

1.20

 walked

1.20

WALK

1.20

Walking

1.18

Activations Density 0.070%

instances of the word "walk" and its variations, indicating a focus on walking actions and activities

No Comments

No Known Activations