INDEX

Explanations

various programming constructs, particularly focusing on code elements and structures

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

 noDo

-0.42

 Lightboxes

-0.39

 Wikimedijinoj

-0.33

期刊论文

-0.33



-0.32

Autoritní

-0.31

UserScript

-0.30

 nahilalakip

-0.29

@"/

-0.29

 Riproduzione

-0.29

POSITIVE LOGITS

DockStyle

0.63

OGND

0.61

AnchorStyles

0.57

writeField

0.54

뀔

0.53

MLLoader

0.53

ArgsConstructor

0.52

LookAnd

0.52

polation

0.51

 oprot

0.51

Activations Density 0.003%

various programming constructs, particularly focusing on code elements and structures

No Comments

No Known Activations