INDEX

Explanations

discussions around moderation and balance in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

Up

-0.49

 changed

-0.44

 debajo

-0.43

Up

-0.42

 name

-0.40

 chief

-0.39

 cambió

-0.38

 gấp

-0.37

up

-0.37

POSITIVE LOGITS

存于互联网档案馆

0.84

 itſelf

0.83

 excès

0.82

 overboard

0.81

 overdo

0.81

 Limits

0.78

Personendaten

0.78

didSet

0.78

enumii

0.77

 limits

0.77

Activations Density 0.320%

discussions around moderation and balance in various contexts

No Comments

No Known Activations