INDEX

Explanations

elements and attributes associated with weaponry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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

ècie

-0.53

Dolly

-0.39

 Narco

-0.38

pushFollow

-0.38

 Dolly

-0.37

 Moles

-0.37

 subsidence

-0.36

NRC

-0.36

Zoo

-0.35

Bom

-0.35

POSITIVE LOGITS

 sword

1.52

 swords

1.43

 Sword

1.27

sword

1.25

Sword

1.20

 Swords

1.19

 Schwert

1.03

 espada

1.00

 blade

0.98

 weapons

0.96

Activations Density 0.505%

elements and attributes associated with weaponry

No Comments

No Known Activations