INDEX

Explanations

quotations within text

oai_token-act-pair · gpt-3.5-turbo

quoted speech or dialogue within the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥĳ

-0.98

etheless

-0.89

~~~~

-0.85

£ı

-0.84

NetMessage

-0.81

DragonMagazine

-0.79

ãĤ¦ãĤ¹

-0.76

soDeliveryDate

-0.76

ĻĤ

-0.75

XXX

-0.74

POSITIVE LOGITS

 said

1.51

said

1.30

 says

1.24

 joked

1.21

 remarked

1.20

 commented

1.14

 exclaimed

1.10

 explained

1.08

 wrote

1.07

 lamented

1.03

Activations Density 0.090%

quotations within text

quoted speech or dialogue within the text

No Comments

No Known Activations