INDEX

Explanations

question marks at the end of sentences

oai_token-act-pair · gpt-3.5-turbo

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

 Strauss

-0.62

 themselves

-0.55

 Bord

-0.55

 seminal

-0.53

 intrinsically

-0.53

 principally

-0.53

 Vaugh

-0.53

 among

-0.52

 widely

-0.52

 loosely

-0.52

POSITIVE LOGITS

 Login

1.25

 Submit

1.12

Login

1.10

ername

1.09

 Cancel

1.08

 login

1.07

 Password

1.06

 Username

1.04

login

1.00

 Please

0.94

Activations Density 0.494%

question marks at the end of sentences

No Comments

No Known Activations