INDEX

Explanations

references to the letter 'g'

oai_token-act-pair · gpt-3.5-turbo

references to nutritional components or values in a text

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

yssey

-0.70

yrights

-0.70

orea

-0.67

starter

-0.65

thia

-0.64

henko

-0.64

addon

-0.63

ylum

-0.63

ayson

-0.63

ima

-0.63

POSITIVE LOGITS

3.14

1.75

1.54

1.33

gh

1.29

gal

1.28

gn

1.28

ga

1.26

gir

1.24

gu

1.21

Activations Density 0.031%

references to the letter 'g'

references to nutritional components or values in a text

No Comments

No Known Activations