INDEX

Explanations

the word "petty" with varying degrees of emphasis

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "petty" 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 @ 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

Not in Any Lists

No Comments

Negative Logits

IBLE

-0.84

icken

-0.81

hetti

-0.81

ources

-0.78

earchers

-0.76

bos

-0.72

 Gutenberg

-0.72

OHN

-0.72

igslist

-0.71

enic

-0.70

POSITIVE LOGITS

 theft

0.88

 petty

0.87

 thief

0.76

 Petty

0.74

 thefts

0.72

 jealous

0.71

 misdemeanor

0.70

 thieves

0.69

 arithmetic

0.68

 duty

0.67

Activations Density 0.018%

the word "petty" with varying degrees of emphasis

occurrences of the word "petty" in various contexts

No Comments

No Known Activations