INDEX

Explanations

mentions of the name "Mario."

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Mario" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aylor

-0.85

ynski

-0.80

arget

-0.74

 Newsp

-0.69

ORE

-0.69

oring

-0.67

ivities

-0.67

ribune

-0.67

naire

-0.65

 warrants

-0.65

POSITIVE LOGITS

 Kart

1.36

 Luigi

1.03

 Mario

0.94

 Bros

0.89

Mario

0.89

 Cuomo

0.84

otti

0.80

 Polo

0.78

Maker

0.77

Wii

0.76

Activations Density 0.013%

mentions of the name "Mario."

mentions of the name "Mario" and its variations in different contexts

No Comments

No Known Activations