INDEX

Explanations

words related to candles or candlelight

oai_token-act-pair · gpt-3.5-turbo

references to candles and torches

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

doms

-0.85

ese

-0.76

ocard

-0.72

ordan

-0.72

cca

-0.71

agara

-0.70

erion

-0.68

FactoryReloaded

-0.68

æĸ¹

-0.67

REDACTED

-0.65

POSITIVE LOGITS

lit

1.06

 candle

1.04

 candles

1.03

 lantern

0.94

 Candle

0.92

 illuminating

0.87

light

0.86

lit

0.85

 illuminated

0.84

vig

0.81

Activations Density 0.058%

words related to candles or candlelight

references to candles and torches

No Comments

No Known Activations