INDEX

Explanations

phrases related to checking facts or telling the truth

oai_token-act-pair · gpt-3.5-turbo

elements related to fact-checking and truth assessment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ebus

-0.93

reciation

-0.78

itsch

-0.75

icz

-0.71

Interstitial

-0.70

aceae

-0.69

Pg

-0.68

ustom

-0.68

iod

-0.66

atri

-0.65

POSITIVE LOGITS

Roe

0.76

 dispro

0.68

 falsehood

0.65

 fiction

0.65

etheless

0.64

çİĭ

0.64

 factual

0.64

matter

0.64

 validity

0.63

 inacc

0.63

Activations Density 0.149%

phrases related to checking facts or telling the truth

elements related to fact-checking and truth assessment

No Comments

No Known Activations