INDEX

Explanations

keywords related to legal proceedings or official hearings

oai_token-act-pair · gpt-3.5-turbo

references to public hearings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

resp

-0.76

ku

-0.73

football

-0.71

sy

-0.71

cer

-0.70

rip

-0.70

heres

-0.69

err

-0.69

fall

-0.69

jc

-0.68

POSITIVE LOGITS

 hearings

1.52

 Hearing

1.21

 hearing

1.00

 Hear

0.97

 "$:/

0.90

 proceedings

0.83

 screenings

0.82

imony

0.81

 meetings

0.80

 panels

0.79

Activations Density 0.010%

keywords related to legal proceedings or official hearings

references to public hearings

No Comments

No Known Activations