INDEX

Explanations

phrases related to recommendations and useful attributes of products or experiences

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

 allegedly

-0.51

ductos

-0.47

뀜

-0.46

 apparently

-0.46

 disponibilité

-0.46

]=>

-0.45

😐

-0.45

隈

-0.45

apparently

-0.44

 DialogInterface

-0.43

POSITIVE LOGITS

hver

0.76

any

0.74

 birthdays

0.74

 picnics

0.70

 anniversaries

0.69

 akár

0.68

 nahilalakip

0.67

Parties

0.66

 your

0.66

 anytime

0.64

Activations Density 0.106%

phrases related to recommendations and useful attributes of products or experiences

No Comments

No Known Activations