INDEX

Explanations

words related to cooking appliances, particularly stoves

oai_token-act-pair · gpt-3.5-turbo

references to stoves and heating elements

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ceed

-0.72

aminer

-0.70

 reneg

-0.67

uality

-0.66

naire

-0.64

ulus

-0.64

Ĥİ

-0.63

ordon

-0.63

 assassination

-0.62

 judicial

-0.62

POSITIVE LOGITS

chest

0.92

fell

0.90

plates

0.84

tered

0.81

tering

0.81

 stove

0.79

itute

0.79

plate

0.79

cot

0.78

 rooms

0.77

Activations Density 0.026%

words related to cooking appliances, particularly stoves

references to stoves and heating elements

No Known Activations