INDEX

Explanations

the word "are" in various contexts

oai_token-act-pair · gpt-3.5-turbo

forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

pedia

-0.85

 Highlights

-0.71

 congr

-0.71

inav

-0.71

 Moons

-0.70

 Waste

-0.69

 Tags

-0.66

 Extend

-0.66

 Grape

-0.65

rates

-0.65

POSITIVE LOGITS

 able

1.17

nt

1.07

 willing

1.05

 unable

1.02

 immune

1.00

 obligated

0.99

 supposed

0.98

 aware

0.94

 unaware

0.93

 capable

0.91

Activations Density 0.436%

the word "are" in various contexts

forms of the verb "to be."

No Comments

No Known Activations