INDEX

Explanations

phrases related to legal actions and documentation, specifically regarding the issuance of warrants

oai_token-act-pair · gpt-3.5-turbo

terms related to legal search warrants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Tycoon

-0.88

 Tiff

-0.77

 Rift

-0.71

minecraft

-0.69

 Chick

-0.69

 PLAY

-0.68

 Clubs

-0.68

 Cater

-0.67

 Survivors

-0.65

 Brands

-0.65

POSITIVE LOGITS

 affidavit

1.14

 authorizing

1.12

 warrant

1.09

 warrants

0.96

 affidav

0.95

 authorization

0.94

 exting

0.90

 issued

0.89

NING

0.88

iffs

0.87

Activations Density 0.038%

phrases related to legal actions and documentation, specifically regarding the issuance of warrants

terms related to legal search warrants

No Comments

No Known Activations