INDEX

Explanations

phrases related to data, statistics, or facts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating ease or simplicity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Contents

-0.77

 whilst

-0.74

Firstly

-0.72

terrorist

-0.70

([

-0.68

(-

-0.65

æĪ¦

-0.64

Evil

-0.63

stab

-0.61

(&

-0.61

POSITIVE LOGITS

 nonprofits

0.80

 worries

0.73

 mentors

0.72

 Jacobs

0.71

 anecd

0.70

toggle

0.70

 Sharma

0.69

 anecdotal

0.68

 volunteers

0.68

 Challenges

0.67

Activations Density 1.076%

phrases related to data, statistics, or facts

phrases indicating ease or simplicity

No Comments

No Known Activations