INDEX

Explanations

structured code comments and documentation within programming languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 GoogleFonts

-0.68

TagMode

-0.61

malloc

-0.57

rotto

-0.55

 ganger

-0.54

 malloc

-0.53

Biblia

-0.53

сшта

-0.53

rrggbb

-0.52

 betweenstory

-0.52

POSITIVE LOGITS

 class

1.52

class

1.28

 classes

1.20

 Class

1.13

 classe

1.06

Class

1.04

 clase

1.03

 subclass

0.97

 className

0.94

 Klasse

0.94

Activations Density 0.445%

structured code comments and documentation within programming languages

No Comments

No Known Activations