INDEX

Explanations

the word "tit" followed by either a number or specific letters to make a complete word or concept

oai_token-act-pair · gpt-3.5-turbo

references to "tit" or related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

insky

-0.70

ero

-0.68

 Logic

-0.67

abi

-0.67

abis

-0.65

eway

-0.65

nsic

-0.64

 defects

-0.64

 Greenwood

-0.64

lier

-0.64

POSITIVE LOGITS

tit

4.05

Tit

1.91

 titan

1.53

TIT

1.39

TIT

1.32

mog

1.20

 garg

1.14

 quint

1.05

 tits

1.04

 magn

0.94

Activations Density 0.027%

the word "tit" followed by either a number or specific letters to make a complete word or concept

references to "tit" or related variations

No Comments

No Known Activations