INDEX

Explanations

descriptions of individuals with positive attributes

oai_token-act-pair · gpt-3.5-turbo

traits and characteristics that describe positive qualities in people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

çīĪ

-0.86

ESA

-0.77

atari

-0.76

soDeliveryDate

-0.75

IRE

-0.75

 Versus

-0.74

LCS

-0.73

IDs

-0.72

imester

-0.72

iatus

-0.72

POSITIVE LOGITS

 witty

1.14

 temperament

1.12

 compassionate

1.11

 charisma

1.10

 demeanor

1.09

 intellect

1.08

 charismatic

1.05

 fearless

1.04

 personality

1.04

 intelligent

1.03

Activations Density 0.575%

descriptions of individuals with positive attributes

traits and characteristics that describe positive qualities in people

No Comments

No Known Activations