INDEX

Explanations

names and titles of political figures

oai_token-act-pair · gpt-3.5-turbo

proper nouns or specific names

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Vera

-0.86

 enthusi

-0.81

Byr

-0.75

LIA

-0.74

 Victoria

-0.73

 phys

-0.73

 COURT

-0.70

SEN

-0.70

 Plaint

-0.69

ãĢĮ

-0.69

POSITIVE LOGITS

ink

1.10

oop

1.07

arp

1.06

ap

1.04

isk

1.04

inkle

1.02

ape

1.02

acks

1.02

ip

1.01

opp

0.99

Activations Density 0.245%

names and titles of political figures

proper nouns or specific names

No Comments

No Known Activations