INDEX

Explanations

references to X-Men characters and related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of the "X-Men" franchise

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

kson

-0.88

sburgh

-0.77

assetsadobe

-0.77

kamp

-0.74

 captcha

-0.74

 Lauder

-0.72

Downloadha

-0.71

getic

-0.69

 TODAY

-0.67

swer

-0.67

POSITIVE LOGITS

posed

1.10

avier

1.05

XXXX

1.02

peria

0.97

yz

0.95

iao

0.94

IX

0.93

 chromosome

0.86

anth

0.83

BOX

0.82

Activations Density 0.025%

references to X-Men characters and related terms

mentions of the "X-Men" franchise

No Comments

No Known Activations