INDEX

Explanations

phrases indicating criticism or disappointment in various contexts

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.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

-0.60

 scuro

-0.53

 credible

-0.52

 affirm

-0.51

 alba

-0.48

 TextAppearance

-0.48

SuppressWarnings

-0.48

 légiti

-0.47

 vierge

-0.47

[`

-0.47

POSITIVE LOGITS

TagMode

0.78

GEBURTSDATUM

0.69

EDEFAULT

0.65

complexContent

0.64

 հղումներ

0.62

Réalisation

0.57

 AspNetCore

0.56

ItemLayout

0.55

phrine

0.53

ruptedException

0.51

Activations Density 0.265%

phrases indicating criticism or disappointment in various contexts

No Comments

No Known Activations