INDEX

Explanations

references to locations or entities including the term "Penn."

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Penn"

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

eer

-0.78

 rall

-0.65

 cerebral

-0.63

eering

-0.62

========

-0.61

 balloons

-0.61

LEASE

-0.61

 willful

-0.60

PRES

-0.60

 Malays

-0.60

POSITIVE LOGITS

sylv

1.48

sylvania

1.33

insula

1.00

iless

0.99

 Penn

0.96

sburg

0.91

Penn

0.90

essee

0.88

olini

0.87

sat

0.87

Activations Density 0.011%

references to locations or entities including the term "Penn."

mentions of the word "Penn"

No Comments

No Known Activations