INDEX

Explanations

actionable instructions or steps often found in lists or guides

oai_token-act-pair · gpt-3.5-turbo

words or phrases suggesting warnings or cautionary themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

kus

-0.73

pi

-0.71

 mosqu

-0.70

leaning

-0.68

tight

-0.68

 intest

-0.64

capacity

-0.64

irtual

-0.63

ldon

-0.62

emetery

-0.62

POSITIVE LOGITS

BOOK

0.94

 Generator

0.84

 ABOUT

0.77

 summar

0.76

itionally

0.74

ues

0.73

çīĪ

0.73

 Explan

0.72

 Description

0.72

ably

0.71

Activations Density 0.233%

actionable instructions or steps often found in lists or guides

words or phrases suggesting warnings or cautionary themes

No Comments

No Known Activations