INDEX

Explanations

Twitter handles

oai_token-act-pair · gpt-3.5-turbo

references to social media handles or usernames

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 varied

-0.73

 alone

-0.71

 Labyrinth

-0.71

 Shelter

-0.70

 striking

-0.69

 grasp

-0.68

 shortage

-0.68

 âĢ¦"

-0.68

bag

-0.68

 veto

-0.68

POSITIVE LOGITS

Writ

1.60

1.46

Reports

1.44

Official

1.41

NBC

1.35

Cod

1.30

Ide

1.29

Twe

1.29

Stud

1.29

Coach

1.27

Activations Density 0.105%

Twitter handles

references to social media handles or usernames

No Comments

No Known Activations