INDEX

Explanations

terms related to experimental procedures and measurements

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

ized

-0.81

ITY

-0.75

naire

-0.73

ially

-0.69

berton

-0.69

ALLY

-0.68

\}}

-0.66

conditionally

-0.66

ओं

-0.65

IZED

-0.65

POSITIVE LOGITS

ings

1.14

able

0.89

ingly

0.82

ability

0.78

 Monfieur

0.75

INGS

0.70

 Shakspeare

0.69

manship

0.68

NameInMap

0.65

downs

0.64

Activations Density 0.963%

terms related to experimental procedures and measurements

No Comments

No Known Activations