INDEX

Explanations

the word "nowhere"

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a lack of presence or absence

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

CVE

-0.83

tein

-0.68

 Reef

-0.64

ylene

-0.64

midt

-0.62

 SetTextColor

-0.60

ULT

-0.60

ilt

-0.60

enko

-0.57

isations

-0.57

POSITIVE LOGITS

 else

1.20

Else

1.03

 Else

0.89

xus

0.82

ensical

0.70

upon

0.67

else

0.66

side

0.65

 Near

0.64

Scot

0.63

Activations Density 0.006%

the word "nowhere"

phrases indicating a lack of presence or absence

No Comments

No Known Activations