INDEX

Explanations

interrogative phrases or questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Normdatei

-0.72

Hozzáférés

-0.70

ernalia

-0.69

WindowConstants

-0.67

OGND

-0.64

equalTo

-0.64

onAttach

-0.63

 leksikon

-0.63

 useRouter

-0.63

providedIn

-0.62

POSITIVE LOGITS

 about

0.80

about

0.69

 dare

0.66

AsUp

0.64

About

0.62

 ever

0.62

 About

0.61

 else

0.56

 bout

0.55

 could

0.54

Activations Density 0.048%

interrogative phrases or questions

No Comments

No Known Activations