INDEX

Explanations

words ending with '-ator' or '-iator'

oai_token-act-pair · gpt-3.5-turbo

words related to various roles, positions, and titles associated with authorship or agency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ITNESS

-0.83

lyak

-0.75

erton

-0.68

lez

-0.68

erest

-0.67

OVER

-0.66

soDeliveryDate

-0.64

MER

-0.62

ness

-0.61

endor

-0.61

POSITIVE LOGITS

ially

1.37

ium

1.21

ial

1.16

SHIP

1.16

hips

1.09

IAL

1.01

IUM

0.95

io

0.91

ials

0.90

iate

0.88

Activations Density 0.143%

words ending with '-ator' or '-iator'

words related to various roles, positions, and titles associated with authorship or agency

No Comments

No Known Activations