INDEX

Explanations

comparative statements indicating superiority

oai_token-act-pair · gpt-3.5-turbo

comparisons that express superiority or preference

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

urther

-0.70

VT

-0.70

ilic

-0.68

ural

-0.65

hran

-0.63

illary

-0.63

Accessory

-0.63

 Juda

-0.62

CLASSIFIED

-0.61

 enthusi

-0.61

POSITIVE LOGITS

abase

0.84

average

0.81

 average

0.76

atever

0.73

 usual

0.73

atos

0.71

ĻĤ

0.68

 averages

0.67

 anything

0.66

 ours

0.64

Activations Density 0.048%

comparative statements indicating superiority

comparisons that express superiority or preference

No Comments

No Known Activations