INDEX

Explanations

lines or sections of code or data input

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

 embaraz

-0.37

avyzd

-0.36

ThemeOverlay

-0.35

 bague

-0.34

 Verenigde

-0.34

charpe

-0.33

insee

-0.31

ględ

-0.31

 dependiendo

-0.31

ufig

-0.31

POSITIVE LOGITS

 resourceCulture

0.69

SourceChecksum

0.69

expandindo

0.58

WebElementEntity

0.57

UnsafeEnabled

0.56

IUrlHelper

0.55

IContainer

0.55

ErrUnexpectedEOF

0.53

 <<<<<<<<<<<<<<

0.52

ScopeManager

0.52

Activations Density 0.000%

lines or sections of code or data input

No Comments

No Known Activations