INDEX

Explanations

calls to action related to helpful information or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

 talaga

-0.42

-0.41

-0.40

вата

-0.40

-0.39

 Cependant

-0.38

ol

-0.38

 however

-0.38

it

-0.37

POSITIVE LOGITS

 Related

1.22

 related

1.16

Related

1.13

related

1.03

expandindo

1.01

 RELATED

0.92

similar

0.91

popular

0.91

tvguidetime

0.89

Personensuche

0.89

Activations Density 0.102%

calls to action related to helpful information or recommendations

No Comments

No Known Activations