INDEX

Explanations

words indicating options or choices

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases indicating alternatives or options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pony

-0.73

Ont

-0.68

arks

-0.68

got

-0.64

 Aires

-0.63

pecting

-0.63

 Hunters

-0.63

eared

-0.63

Ļ

-0.62

 Wester

-0.62

POSITIVE LOGITS

 postpone

1.11

 negate

1.07

 abolish

1.06

 discontin

1.05

 eliminate

1.03

 cancel

1.02

 delay

1.02

 defund

1.00

 omit

1.00

 disable

0.95

Activations Density 0.119%

words indicating options or choices

conjunctions and phrases indicating alternatives or options

No Comments

No Known Activations