INDEX

Explanations

phrases related to a person named "Loreta" or "Lore."

oai_token-act-pair · gpt-3.5-turbo

references to individuals named Loretta or variations of the name Lore

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

akra

-0.86

ixties

-0.71

ruciating

-0.68

obs

-0.68

uyomi

-0.66

inished

-0.66

obar

-0.65

ertility

-0.64

 Mehran

-0.64

icago

-0.63

POSITIVE LOGITS

tta

1.12

nz

1.10

 Lore

1.08

lette

0.97

tto

0.96

walker

0.92

bolt

0.90

gon

0.90

nce

0.87

le

0.86

Activations Density 0.008%

phrases related to a person named "Loreta" or "Lore."

references to individuals named Loretta or variations of the name Lore

No Comments

No Known Activations