INDEX

Explanations

phrases related to adding items or elements to something

oai_token-act-pair · gpt-3.5-turbo

phrases related to adding or modifying elements in a system or project

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

zee

-0.76

yah

-0.75

Present

-0.73

brain

-0.71

«

-0.68

Building

-0.68

present

-0.67

hat

-0.64

ongyang

-0.64

grim

-0.64

POSITIVE LOGITS

endum

1.20

 insult

1.07

itions

1.03

itional

0.97

ition

0.96

ictions

0.93

resses

0.89

ictive

0.87

 additional

0.83

icts

0.82

Activations Density 0.056%

phrases related to adding items or elements to something

phrases related to adding or modifying elements in a system or project

No Comments

No Known Activations