INDEX

Explanations

proper nouns, particularly names of people and places

oai_token-act-pair · gpt-3.5-turbo

names of individuals or references to specific entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ivity

-0.90

ivism

-0.76

istics

-0.72

ional

-0.66

draw

-0.64

iger

-0.62

mond

-0.62

ourcing

-0.61

acters

-0.61

×Ļ×

-0.61

POSITIVE LOGITS

ONEY

0.85

ORPG

0.83

afia

0.78

 Mayhem

0.76

hattan

0.76

unicip

0.75

onduct

0.73

ichael

0.72

useum

0.69

unction

0.69

Activations Density 0.257%

proper nouns, particularly names of people and places

names of individuals or references to specific entities

No Comments

No Known Activations