INDEX

Explanations

information related to a specific company

oai_token-act-pair · gpt-3.5-turbo

the acronym "PL" followed by a numerical indication, likely related to financial or regulatory entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Mik

-0.72

 Inquisitor

-0.68

 Aber

-0.67

 beef

-0.62

cac

-0.62

 cast

-0.60

 castles

-0.60

haz

-0.60

 adjourn

-0.60

hay

-0.60

POSITIVE LOGITS

PL

4.02

PLA

2.12

PLE

1.94

PL

1.91

PLIC

1.67

pl

1.64

PS

1.50

PD

1.46

PR

1.46

SL

1.46

Activations Density 0.020%

information related to a specific company

the acronym "PL" followed by a numerical indication, likely related to financial or regulatory entities

No Comments

No Known Activations