INDEX

Explanations

percentages and their representations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 obstante

-0.52

 relazioni

-0.48

 vectorielle

-0.48

 calitate

-0.44

Erreferentziak

-0.44

 tartalomajánló

-0.43

 trône

-0.43

 renou

-0.43

 تضيفلها

-0.43

 estekak

-0.42

POSITIVE LOGITS

 thanks

0.66

 courtesy

0.66

 '\\;'

0.64

net

0.60

courtesy

0.56

 plus

0.56

 mere

0.53

 percentage

0.51

percentage

0.48

 grâce

0.47

Activations Density 0.104%

percentages and their representations

No Comments

No Known Activations