INDEX

Explanations

specific symbols or characters

oai_token-act-pair · gpt-3.5-turbo

instances of significant incidents or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 branches

-0.75

 luck

-0.75

 altern

-0.74

 machine

-0.73

 ridic

-0.71

 myself

-0.69

 deduct

-0.69

 retard

-0.68

 scales

-0.68

 additionally

-0.67

POSITIVE LOGITS

Updated

1.42

Published

1.37

Posted

1.36

Enlarge

1.25

ccording

1.24

Guest

1.23

TOR

1.23

WASHINGTON

1.20

LOS

1.18

By

1.18

Activations Density 0.346%

specific symbols or characters

instances of significant incidents or events

No Comments

No Known Activations