INDEX

Explanations

titles or credits in written text

oai_token-act-pair · gpt-3.5-turbo

sentence-ending punctuation

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

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

 orderly

-0.74

 defin

-0.72

 manif

-0.71

 disadvant

-0.71

 shutting

-0.71

 starvation

-0.69

assic

-0.68

 suspic

-0.68

 nuts

-0.66

 physiology

-0.66

POSITIVE LOGITS

 Subscribe

1.32

 Follow

1.21

 Readers

1.12

 Become

1.11

 Visit

1.07

 Please

1.04

Follow

1.03

 Previously

1.00

 Contact

0.99

 Copyright

0.99

Activations Density 0.114%

titles or credits in written text

sentence-ending punctuation

No Comments

No Known Activations