INDEX

Explanations

repeated phrases and references to actions, particularly those beginning with "to."

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

jenigen

-0.54

lapsingToolbar

-0.51

ьере

-0.51

loem

-0.50

jenige

-0.48

 חיצוניים

-0.47

 angekommen

-0.46

jaros

-0.45

ⓘ

-0.45

 akrab

-0.45

POSITIVE LOGITS

 perfection

0.85

 death

0.83

 فريبيس

0.75

 propOrder

0.70

death

0.69

 pieces

0.65

 DEATH

0.63

 shreds

0.63

 Seeder

0.63

 distraction

0.62

Activations Density 0.164%

repeated phrases and references to actions, particularly those beginning with "to."

No Comments

No Known Activations