INDEX

Explanations

various forms of the word "decorate" and related terms

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

Stateless

-0.48

 Autorisations

-0.47

 endforeach

-0.47

ClientRect

-0.46

PointerException

-0.44

 Sigmund

-0.42

VolleyError

-0.42

readFileSync

-0.41

 Bacteriol

-0.41

IPv

-0.41

POSITIVE LOGITS

 decoration

0.90

 Decoration

0.87

 decorations

0.82

 decorating

0.79

 decorate

0.79

 Decor

0.76

 decorated

0.74

 Decorations

0.73

decor

0.73

Decoration

0.72

Activations Density 0.272%

various forms of the word "decorate" and related terms

No Comments

No Known Activations