INDEX

Explanations

phrases indicating strong emotional responses or relationships

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

expandindo

-0.65

NECT

-0.55

LError

-0.54

tonsoft

-0.53

ArrowToggle

-0.53

ETHING

-0.53

(__('

-0.52

HideFlags

-0.51

"¿

-0.51

gehende

-0.49

POSITIVE LOGITS

 ویکی‌پدیا

0.83

XmlAccessorType

0.81

 évêque

0.71

MENAFN

0.68

”—

0.66

 église

0.64

 venuto

0.63

”.

0.62

 conduite

0.61

 undersigned

0.61

Activations Density 0.290%

phrases indicating strong emotional responses or relationships

No Comments

No Known Activations