INDEX

Explanations

terms and phrases related to safety measures and standards

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

adeloupe

-0.76

 Jackman

-0.74

TypeDef

-0.69

 bint

-0.66

 direta

-0.66

 maș

-0.66

󠁢

-0.65

mik

-0.65

 république

-0.65

 isolado

-0.64

POSITIVE LOGITS

afety

1.07

 Safety

1.06

 safety

1.05

Safety

1.05

SAFETY

0.91

 SAFETY

0.91

Security

0.85

 security

0.84

safety

0.81

enumi

0.80

Activations Density 0.068%

terms and phrases related to safety measures and standards

No Comments

No Known Activations