INDEX

Explanations

names of individuals and locations

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Glad."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ccording

-0.81

externalActionCode

-0.78

uador

-0.74

BOOK

-0.74

perty

-0.70

merce

-0.68

lished

-0.68

recorded

-0.67

ð

-0.66

anwhile

-0.65

POSITIVE LOGITS

 Glad

1.25

Tid

0.98

well

0.95

igue

0.85

imir

0.84

itional

0.83

ifiers

0.78

endar

0.73

rill

0.72

DeL

0.72

Activations Density 0.007%

names of individuals and locations

mentions of the name "Glad."

No Comments

No Known Activations