INDEX

Explanations

keywords related to applications or requests for action

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "App" in various contexts

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

terday

-0.88

ovych

-0.71

 outdoors

-0.66

 cigars

-0.66

 knot

-0.65

 methyl

-0.64

 Mond

-0.63

warts

-0.62

EngineDebug

-0.61

EStreamFrame

-0.61

POSITIVE LOGITS

ointment

1.41

alach

1.41

ropri

1.32

earance

1.30

reciation

1.30

ortion

1.29

earances

1.22

rehens

1.21

endix

1.21

rentices

1.19

Activations Density 0.021%

keywords related to applications or requests for action

mentions of the word "App" in various contexts

No Comments

No Known Activations