INDEX

Explanations

instances of coercion or compulsion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

subsid

-0.48

 Subsid

-0.46

 SWEET

-0.45

 subsidy

-0.43

utilisons

-0.42

 plainte

-0.42

theft

-0.42

sweet

-0.41

 theft

-0.40

setAge

-0.40

POSITIVE LOGITS

 forced

1.40

 force

1.38

 forcing

1.38

force

1.30

 forces

1.25

 compelled

1.21

 Force

1.18

forced

1.18

 Forced

1.17

Force

1.16

Activations Density 0.206%

instances of coercion or compulsion

No Comments

No Known Activations