INDEX

Explanations

words related to scanning and digital imaging

oai_token-act-pair · gpt-3.5-turbo

instances of the word "scan" and related variations

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

 impunity

-0.71

 Cind

-0.69

soDeliveryDate

-0.63

emort

-0.62

doms

-0.61

orney

-0.60

OTAL

-0.60

soType

-0.60

Aid

-0.60

idences

-0.59

POSITIVE LOGITS

 scans

1.12

lator

1.11

 scan

1.07

 scanning

0.95

lations

0.90

 scanners

0.88

 Scan

0.86

ning

0.85

 scanned

0.81

 scanner

0.79

Activations Density 0.014%

words related to scanning and digital imaging

instances of the word "scan" and related variations

No Comments

No Known Activations