INDEX

Explanations

elements related to 'SN' or 'Leslie' in the text

oai_token-act-pair · gpt-3.5-turbo

references to "SN" or similar abbreviations, possibly related to scientific or technical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

fleet

-0.70

quez

-0.67

icum

-0.66

Corp

-0.66

 Horus

-0.65

ordon

-0.65

inus

-0.65

Marcus

-0.64

esian

-0.64

fax

-0.64

POSITIVE LOGITS

SN

3.96

SN

2.60

NES

1.46

 SNAP

1.41

SNP

1.29

NTS

1.25

KN

1.23

TNT

1.22

SM

1.20

NES

1.19

Activations Density 0.021%

elements related to 'SN' or 'Leslie' in the text

references to "SN" or similar abbreviations, possibly related to scientific or technical contexts

No Comments

No Known Activations