INDEX

Explanations

occurrences of the word "und"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 autorytatywna

-0.84

 houſe

-0.69

 ejus

-0.68

 aéri

-0.64

 atve

-0.63

 religieuse

-0.62

 Stiles

-0.62

ře

-0.62

 skolan

-0.61

 stră

-0.61

POSITIVE LOGITS

und

2.10

Und

1.65

UND

1.51

Und

1.48

und

1.46

 oder

1.22

UND

1.13

och

0.95

og

0.87

tund

0.87

Activations Density 0.075%

occurrences of the word "und"

No Comments

No Known Activations