INDEX

Explanations

names and locations

oai_token-act-pair · gpt-3.5-turbo

abbreviations, names, and specific identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 unequ

-0.69

 tails

-0.60

gallery

-0.60

Indian

-0.60

paper

-0.56

creen

-0.56

ãĢĲ

-0.55

 toget

-0.54

aminer

-0.54

Paper

-0.53

POSITIVE LOGITS

isse

0.76

illard

0.73

ONSORED

0.71

uez

0.71

ocus

0.71

 Madness

0.70

ibles

0.69

arson

0.69

ement

0.68

uve

0.67

Activations Density 0.277%

names and locations

abbreviations, names, and specific identifiers

No Comments

No Known Activations