INDEX

Explanations

explicitly stated acknowledgments or declarations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "formally" and related formal declarations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

=-=-=-=-=-=-=-=-

-0.81

rug

-0.72

 Will

-0.71

Wat

-0.70

 Eden

-0.69

onal

-0.67

Pages

-0.67

 Perse

-0.66

 Belt

-0.66

 Gore

-0.66

POSITIVE LOGITS

 formally

1.14

 formal

0.96

 induct

0.93

 exting

0.90

 satell

0.84

 disav

0.80

comings

0.80

 Lumpur

0.80

 lodged

0.80

ntil

0.79

Activations Density 0.004%

explicitly stated acknowledgments or declarations

instances of the word "formally" and related formal declarations

No Comments

No Known Activations