INDEX

Explanations

mentions of things or activities that can be enjoyed

oai_token-act-pair · gpt-3.5-turbo

the word "enjoy" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amm

-0.72

ural

-0.71

Wr

-0.67

ene

-0.65

voy

-0.65

metal

-0.64

iop

-0.64

cci

-0.63

orne

-0.63

container

-0.62

POSITIVE LOGITS

 enjoy

1.07

 enjoying

1.05

 Enjoy

1.02

 enjoyed

0.99

 enjoys

0.98

 enjoyment

0.97

lihood

0.96

Enjoy

0.87

joy

0.84

nels

0.83

Activations Density 0.013%

mentions of things or activities that can be enjoyed

the word "enjoy" and its variations in different contexts

No Comments

No Known Activations