INDEX

Explanations

references to numerical values or counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 CURIAM

-0.59

RESUMO

-0.56

 peores

-0.55

chappement

-0.54

 épais

-0.54



-0.54

omiast

-0.53

cetamol

-0.53

 defStyle

-0.53

yelidikan

-0.53

POSITIVE LOGITS

 number

1.21

 NUMBER

1.00

 Number

0.96

исленность

0.94

number

0.91

 amount

0.89

 número

0.84

Number

0.83

NUMBER

0.83

 numberOf

0.81

Activations Density 0.052%

references to numerical values or counts

No Comments

No Known Activations