INDEX

Explanations

quotes and direct speech

oai_token-act-pair · gpt-3.5-turbo

words or phrases that indicate reported speech or quotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

¥µ

-0.76

illance

-0.75

cci

-0.73

iamond

-0.72

BAT

-0.72

 pouch

-0.70

 beads

-0.69

ara

-0.69

£ı

-0.69

ecast

-0.69

POSITIVE LOGITS

Fre

2.48

Fre

2.46

 Freeman

2.02

fre

1.84

fre

1.59

FRE

1.52

 Freak

1.50

Gre

1.40

 Frey

1.22

Gre

1.14

Activations Density 0.231%

quotes and direct speech

words or phrases that indicate reported speech or quotations

No Comments

No Known Activations