INDEX

Explanations

phrases indicating direct speech or reported speech

oai_token-act-pair · gpt-3.5-turbo

dialogue and reported speech in 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¼

-0.89

spir

-0.85

wcs

-0.84

 interfered

-0.84

ãĥİ

-0.84

successfully

-0.82

Login

-0.81

endars

-0.81

Authent

-0.80

{"

-0.78

POSITIVE LOGITS

Mr

1.04

Ms

1.02

Dr

0.98

 Diane

0.97

 Valerie

0.96

 Barbara

0.95

 Jennifer

0.95

Sgt

0.95

 Stephanie

0.95

Joe

0.94

Activations Density 0.056%

phrases indicating direct speech or reported speech

dialogue and reported speech in text

No Comments

No Known Activations