INDEX

Explanations

connective words that indicate relationships or combinations between ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

enumi

-0.53

 skimage

-0.49

'',

-0.46

 BoxDecoration

-0.46

OnTop

-0.43

ابی

-0.42

":"",

-0.41

meneu

-0.40

 hereof

-0.40

brano

-0.40

POSITIVE LOGITS

 autorytatywna

0.99

etc

0.96

IntoConstraints

0.94

 persino

0.94

etc

0.90

 even

0.86

 tartalomajánló

0.82

 addirittura

0.79

 zelfs

0.79

بوابة

0.77

Activations Density 0.256%

connective words that indicate relationships or combinations between ideas

No Comments

No Known Activations