INDEX

Explanations

mentions of the retailer "Wal-Mart"

oai_token-act-pair · gpt-3.5-turbo

mentions of "Wal-Mart" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Carbuncle

-0.77

INF

-0.68

 familiar

-0.65

 habitable

-0.64

 appropriation

-0.64

 Darkness

-0.63

 leukemia

-0.62

IBLE

-0.62

 Blow

-0.59

 Immunity

-0.58

POSITIVE LOGITS

nut

1.35

nuts

1.20

cott

1.18

halla

1.05

Mart

1.04

mart

1.01

rus

0.98

pole

0.97

ther

0.95

ston

0.93

Activations Density 0.013%

mentions of the retailer "Wal-Mart"

mentions of "Wal-Mart" in various contexts

No Comments

No Known Activations