INDEX

Explanations

variations of ordinal numbers and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 pleaſure

-0.80

awtextra

-0.79

engertian

-0.75

seamnă

-0.73

createCanvas

-0.72

 fufficient

-0.72

 houſe

-0.71

 perfons

-0.70

RectangleBorder

-0.69

 ſhe

-0.69

POSITIVE LOGITS

op

0.80

 betweenstory

0.79

las

0.77

al

0.71

las

0.70

tonsoft

0.67

vor

0.63

ve

0.63

lid

0.62

í

0.62

Activations Density 0.111%

variations of ordinal numbers and related terms

No Comments

No Known Activations