INDEX

Explanations

geometric attributes of shapes

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

ambién

-0.64

 <>",

-0.58

piecze

-0.56

AndEndTag

-0.54

Hochspringen

-0.53

 Accesat

-0.51

 iprot

-0.50

QApplication

-0.49

Vegetarian

-0.48

发表于

-0.48

POSITIVE LOGITS

 BorderRadius

1.24

InputBorder

0.49

 borderRadius

0.46

RRect

0.43

borderRadius

0.42

Kanpo

0.39

Personendaten

0.38

OGND

0.36

 currentNode

0.36

 AssemblyVersion

0.35

Activations Density 0.000%

geometric attributes of shapes

No Comments

No Known Activations