INDEX

Explanations

actions related to removal or deletion processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

UnsafeEnabled

-0.45

 ulang

-0.40

 skipping

-0.38

 demise

-0.38

 downfall

-0.36

 transférez

-0.36

different

-0.36

 propOrder

-0.36

 ivelany

-0.36

Scénario

-0.36

POSITIVE LOGITS

 unnecessary

1.16

 unwanted

1.11

 excess

1.02

 superfluous

1.00

 redundant

0.95

 inneces

0.93

excess

0.92

 лиш

0.87

 extraneous

0.86

Excess

0.84

Activations Density 0.743%

actions related to removal or deletion processes

No Comments

No Known Activations