INDEX

Explanations

references to menu items and food dishes

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

conomie

-0.44

ifflin

-0.41

sci

-0.41

dchen

-0.41

zeß

-0.40

ándalo

-0.40

viaf

-0.40

 machete

-0.40

mpara

-0.39

 thee

-0.39

POSITIVE LOGITS

sand

0.57

biscuit

0.54

rose

0.51

rain

0.50

flower

0.50

ooze

0.50

wood

0.49

cucumber

0.49

bamboo

0.48

ice

0.48

Activations Density 0.732%

references to menu items and food dishes

No Comments

No Known Activations