INDEX

Explanations

references to political figures and endorsements or statements about them

oai_token-act-pair · gpt-3.5-turbo

phrases associated with names and their significance

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

olution

-0.75

olved

-0.64

olutions

-0.63

ombo

-0.63

processing

-0.62

ruction

-0.61

Parameter

-0.60

Abstract

-0.60

Storage

-0.59

apter

-0.59

POSITIVE LOGITS

his

1.08

His

1.00

his

0.97

he

0.95

His

0.92

He

0.76

He

0.76

vou

0.76

 temperament

0.73

 himself

0.73

Activations Density 1.098%

references to political figures and endorsements or statements about them

phrases associated with names and their significance

No Comments

No Known Activations