INDEX

Explanations

references to injuries and wounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 tartalomajánló

-0.57

 propOrder

-0.57

RegistryLite

-0.54

textAppearance

-0.54

曖昧さ回避

-0.53

 levure

-0.51

herr

-0.47

 ума

-0.47

rinol

-0.47

onsored

-0.47

POSITIVE LOGITS

 wounds

1.48

 wound

1.38

 Wounds

1.30

 Wound

1.29

Wound

1.28

wound

1.28

 wounded

1.20

 injury

1.15

 Injury

1.14

Injury

1.14

Activations Density 0.501%

references to injuries and wounds

No Comments

No Known Activations