INDEX

Explanations

programming terminology and concepts related to types and structures in code

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

AndEndTag

-0.74

Diwedd

-0.71

IsMutable

-0.70

ArgsConstructor

-0.70

MessageOf

-0.69

<unused41>

-0.64

<unused79>

-0.64

<unused14>

-0.64

 好文分享

-0.64

<unused42>

-0.64

POSITIVE LOGITS

app

0.31

ţe

0.31

0.29

Lea

0.29

Or

0.28

 apapun

0.28

țe

0.28

the

0.27

 constraints

0.26

 вкусу

0.26

Activations Density 0.026%

programming terminology and concepts related to types and structures in code

No Comments

No Known Activations