INDEX

Explanations

variations of the word "Works"

oai_token-act-pair · gpt-3.5-turbo

references to various entities or organizations that include "Works" in their name

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

SPONSORED

-0.75

Cath

-0.67

 Adin

-0.64

ORN

-0.63

----------

-0.62

antha

-0.60

ANA

-0.60

 Ying

-0.60

arta

-0.60

clad

-0.59

POSITIVE LOGITS

hops

1.59

paces

1.50

heet

1.23

pace

1.20

hirt

1.03

hift

1.01

icle

0.96

icles

0.92

afe

0.91

esian

0.87

Activations Density 0.040%

variations of the word "Works"

references to various entities or organizations that include "Works" in their name

No Comments

No Known Activations