INDEX

Explanations

elements of structured lists or format indicators in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

bewerken

-0.84

 ***!

-0.80

UnusedPrivate

-0.77

 henvisninger

-0.75

AndEndTag

-0.73

Obrázky

-0.70

 виправивши

-0.69

 numerus

-0.68

 ddelweddau

-0.68

 nakalista

-0.66

POSITIVE LOGITS

stretchr

0.45

 autant

0.42

sträger

0.42

any

0.41

теристики

0.41

 contacter

0.40

ibilité

0.40

掉的

0.40

 adorno

0.40

or

0.39

Activations Density 0.813%

elements of structured lists or format indicators in a document

No Comments

No Known Activations