INDEX

Explanations

quotations within text

oai_token-act-pair · gpt-3.5-turbo

quotations or quotation marks

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

 Archdemon

-0.88

Ari

-0.83

oun

-0.83

Iv

-0.82

 accomp

-0.82

 Alban

-0.81

 addresses

-0.80

ads

-0.80

 administrators

-0.79

Arn

-0.79

POSITIVE LOGITS

massive

1.63

huge

1.61

significant

1.59

major

1.54

small

1.53

classic

1.52

serious

1.52

very

1.51

little

1.51

special

1.51

Activations Density 0.035%

quotations within text

quotations or quotation marks

No Comments

No Known Activations