INDEX

Explanations

instances of the word "guide" followed by a noun or verb

oai_token-act-pair · gpt-3.5-turbo

references to guides and guidebooks

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

 concess

-0.77

 spect

-0.70

 ILCS

-0.70

asted

-0.67

ammy

-0.67

ntil

-0.63

 Mellon

-0.62

oubted

-0.62

 forfeit

-0.62

yss

-0.61

POSITIVE LOGITS

book

1.39

books

1.32

guide

1.09

posts

1.09

BOOK

1.07

 guide

0.89

tips

0.85

tip

0.84

Book

0.84

kit

0.83

Activations Density 0.038%

instances of the word "guide" followed by a noun or verb

references to guides and guidebooks

No Comments

No Known Activations