INDEX

Explanations

phrases related to supporting or confirming something, such as upholding laws or principles

oai_token-act-pair · gpt-3.5-turbo

terms related to maintaining or reinforcing laws and principles

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

NetMessage

-0.84

apter

-0.83

goo

-0.75

ascus

-0.66

ohydrate

-0.66

bor

-0.65

pox

-0.65

ãĤ¼ãĤ¦ãĤ¹

-0.64

 explode

-0.64

borgh

-0.63

POSITIVE LOGITS

ably

1.01

 upholding

0.88

enance

0.84

 upheld

0.79

ORED

0.73

 principles

0.71

 consistency

0.71

ively

0.71

cies

0.71

 uphold

0.70

Activations Density 0.021%

phrases related to supporting or confirming something, such as upholding laws or principles

terms related to maintaining or reinforcing laws and principles

No Comments

No Known Activations