INDEX

Explanations

references to numbers and their formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 demais

-0.83

 initComponents

-0.81

Tembelea

-0.77

 Borja

-0.77

 pObj

-0.73

 leaſt

-0.73

 blest

-0.72

"]();

-0.71

bbene

-0.71

"\<

-0.70

POSITIVE LOGITS

 number

1.75

 Number

1.64

 numbers

1.58

 NUMBER

1.58

Number

1.53

number

1.46

NUMBER

1.43

 Numbers

1.41

Numbers

1.34

numbers

1.33

Activations Density 0.119%

references to numbers and their formatting

No Comments

No Known Activations