INDEX

Explanations

phrases describing something as foolish or lacking intelligence

oai_token-act-pair · gpt-3.5-turbo

instances of the word "dumb" in varying contexts and usages

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

Lago

-0.82

OHN

-0.73

 Noir

-0.73

AUT

-0.68

ATURES

-0.68

APH

-0.67

-0.66

UAL

-0.65

akings

-0.65

uberty

-0.65

POSITIVE LOGITS

founded

1.24

found

1.03

arton

1.02

bell

0.99

est

0.97

wallet

0.95

stru

0.94

asses

0.92

holes

0.91

 dumb

0.90

Activations Density 0.011%

phrases describing something as foolish or lacking intelligence

instances of the word "dumb" in varying contexts and usages

No Comments

No Known Activations