INDEX

Explanations

patterns of precedent and similarity in actions or ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 chi̍t

-0.50

omé

-0.49

ftant

-0.48

popper

-0.47

kauf

-0.47

stående

-0.47

Sh

-0.47

lard

-0.46

AHAN

-0.46

 Grade

-0.46

POSITIVE LOGITS

 Normdatei

0.71

RenderAtEndOf

0.63

Someone

0.63

rungsseite

0.61

 HasFactory

0.61

 buddha

0.58

kdo

0.57

 Someone

0.57

rrggbb

0.56

 SOMEONE

0.56

Activations Density 0.341%

patterns of precedent and similarity in actions or ideas

No Comments

No Known Activations