INDEX

Explanations

Twitter usernames preceded by an "@" symbol

oai_token-act-pair · gpt-3.5-turbo

Twitter handles or mentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 buildup

-0.86

 prolifer

-0.81

 succeeding

-0.78

 deficient

-0.77

 cooperative

-0.77

 congreg

-0.75

 concluding

-0.75

 unsuccessful

-0.75

 remake

-0.75

 tasting

-0.74

POSITIVE LOGITS

realDonaldTrump

1.77

#$

1.47

eric

1.31

netflix

1.30

john

1.29

jon

1.29

gmail

1.27

dan

1.26

christ

1.24

thereal

1.24

Activations Density 0.038%

Twitter usernames preceded by an "@" symbol

Twitter handles or mentions

No Comments

No Known Activations