INDEX

Explanations

terms related to organizational programs and initiatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 gants

-0.80

 Stearns

-0.78

gameId

-0.74

Pew

-0.69

 boissons

-0.69

 referenties

-0.66

 disks

-0.64

 Abend

-0.64

 Schilder

-0.63

zaj

-0.63

POSITIVE LOGITS

 moulds

1.39

 fibres

1.38

 licences

1.35

 programmes

1.33

Fibre

1.26

 artefacts

1.26

 Vapour

1.25

 Travellers

1.25

 metres

1.24

 favourites

1.24

Activations Density 0.254%

terms related to organizational programs and initiatives

No Comments

No Known Activations