INDEX

Explanations

mentions and descriptions related to events and reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

лтемелер

-0.59

 ModelExpression

-0.56

كويكب

-0.53

oa̍t

-0.53

NameInMap

-0.52

msgTypes

-0.51

HasAnnotation

-0.49

lidean

-0.49

FTFY

-0.48

 мәкал

-0.48

POSITIVE LOGITS

 which

0.55

tagHelperRunner

0.47

ผู้

0.45

 whose

0.45

 selaku

0.44

who

0.43

whose

0.39

 الذي

0.38

 wezen

0.37

 considered

0.37

Activations Density 0.392%

mentions and descriptions related to events and reports

No Comments

No Known Activations