INDEX

Explanations

terms related to avoiding or preventing negative outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

oa̍t

-0.60

 nonUne

-0.55

bootstrapcdn

-0.55

 transfieras

-0.52

:✨

-0.50

 HasFactory

-0.50

AsUp

-0.50

Jus

-0.49

BeginInit

-0.49

Jus

-0.48

POSITIVE LOGITS

 losing

0.47

 risk

0.44

 Verlust

0.44

 loss

0.44

 avoidable

0.38

 typelib

0.36

 crowding

0.36

 Arkivert

0.36

 excess

0.35

 loosing

0.35

Activations Density 0.352%

terms related to avoiding or preventing negative outcomes

No Comments

No Known Activations