INDEX

Explanations

the word "or" preceding contrasting ideas or options

oai_token-act-pair · gpt-3.5-turbo

phrases that present alternatives or choices

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

aturday

-0.70

 Pony

-0.69

ENE

-0.68

ARA

-0.66

EF

-0.65

oft

-0.64

pire

-0.63

enko

-0.63

amon

-0.62

iper

-0.62

POSITIVE LOGITS

 otherwise

1.09

 alternatively

1.05

acles

1.04

nam

1.00

 else

1.00

chard

0.99

Else

0.99

chid

0.98

acle

0.96

acular

0.85

Activations Density 0.094%

the word "or" preceding contrasting ideas or options

phrases that present alternatives or choices

No Comments

No Known Activations