INDEX

Explanations

references to images or photographs being mentioned

oai_token-act-pair · gpt-3.5-turbo

elements related to media and public perception

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

LG

-0.98

 hurd

-0.85

CET

-0.83

atur

-0.82

olars

-0.81

 redist

-0.80

Ã¼r

-0.80

 glim

-0.79

Î

-0.78

LG

-0.77

POSITIVE LOGITS

Joe

2.22

Joe

2.03

 Biden

1.63

jo

1.37

 Joey

1.37

 Joseph

1.33

Joseph

1.23

 Russo

1.13

 Marino

1.00

 Quinn

0.98

Activations Density 0.155%

references to images or photographs being mentioned

elements related to media and public perception

No Comments

No Known Activations