INDEX

Explanations

credit mentions in text

oai_token-act-pair · gpt-3.5-turbo

mentions of attribution or credit in texts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fog

-0.68

 Kinnikuman

-0.68

Kus

-0.65

 pacif

-0.65

 Lans

-0.62

 postwar

-0.62

 acoustic

-0.60

ãĥ¢

-0.60

 marsh

-0.60

Osw

-0.60

POSITIVE LOGITS

 Credit

1.41

Credit

1.39

 Credits

1.07

credit

1.06

 credit

0.97

 CARD

0.92

 Cards

0.91

 Card

0.87

Card

0.86

Credits

0.85

Activations Density 0.006%

credit mentions in text

mentions of attribution or credit in texts

No Comments

No Known Activations