INDEX

Explanations

references to active engagement or participation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 mitään

-0.44

urilor

-0.43

 meille

-0.42

 nantinya

-0.41

 mijne

-0.40

 étr

-0.39

 Anſ

-0.39

 espelho

-0.38

 harganya

-0.38

 espejos

-0.38

POSITIVE LOGITS

 active

1.88

active

1.75

Active

1.71

 Active

1.68

 ACTIVE

1.63

 aktif

1.49

 actives

1.45

 actively

1.44

ACTIVE

1.43

 актив

1.41

Activations Density 0.148%

references to active engagement or participation in various contexts

No Comments

No Known Activations