INDEX

Explanations

names and terms in a foreign language (French and Romanian)

oai_token-act-pair · gpt-3.5-turbo

non-standard characters or symbols

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

oids

-0.88

apsed

-0.79

idious

-0.73

oidal

-0.70

 Eclipse

-0.68

dfx

-0.68

APS

-0.68

enegger

-0.67

othal

-0.65

oid

-0.65

POSITIVE LOGITS

âĤ¬

1.09

Ã¢

0.95

´

0.95

¸

0.94

tre

0.93

\\\\

0.91

Â¯

0.89

Â¯Â¯

0.89

ĺ

0.88

ternity

0.87

Activations Density 0.007%

names and terms in a foreign language (French and Romanian)

non-standard characters or symbols

No Comments

No Known Activations