INDEX

Explanations

significant numerical changes or variations in data

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

DebuggerNonUser

-0.65

ReusableCell

-0.63

SequentialGroup

-0.59

PropertyChanging

-0.56

 referenties

-0.55

AsUp

-0.52

 ujednoznacz

-0.51

InvalidProtocol

-0.50

BeginContext

-0.49

йтинг

-0.49

POSITIVE LOGITS

 significant

0.84

 changes

0.83

 drastic

0.81

 slight

0.78

 rise

0.77

 changements

0.73

 decline

0.73

 increase

0.73

 dramatic

0.71

 substantial

0.71

Activations Density 0.741%

significant numerical changes or variations in data

No Comments

No Known Activations