INDEX

Explanations

mentions of specific technical terms and programming-related words

oai_token-act-pair · gpt-3.5-turbo

references to restoration and reuse practices

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

jury

-0.65

gow

-0.65

waukee

-0.64

ipeg

-0.63

Jaw

-0.62

 Lumpur

-0.60

edin

-0.58

isSpecialOrderable

-0.57

ricanes

-0.57

owitz

-0.56

POSITIVE LOGITS

taboola

0.65

pmwiki

0.64

 Palest

0.63

 disguise

0.60

sic

0.58

catentry

0.57

ĨĴ

0.57

oult

0.56

Mach

0.56

 Artemis

0.55

Activations Density 1.372%

mentions of specific technical terms and programming-related words

references to restoration and reuse practices

No Comments

No Known Activations