INDEX

Explanations

prepositions and conjunctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

__*/

-0.78

RectangleBorder

-0.73

évaluateur

-0.71

 Normdatei

-0.70

AndEndTag

-0.67

 estekak

-0.64

Życiorys

-0.62

 насеље

-0.61

exitRule

-0.61

новниш

-0.61

POSITIVE LOGITS

saraba

0.60

 Consultez

0.55

csolódó

0.54

 recognizing

0.48

ňuje

0.48

 dotyczące

0.47

 recognising

0.46

izr

0.45

 دلیل

0.45

 considering

0.45

Activations Density 1.498%

prepositions and conjunctions

No Comments

No Known Activations