INDEX

Explanations

keywords related to knowledge, identification, determination, and inference

oai_token-act-pair · gpt-3.5-turbo

phrases related to the processes of knowledge verification and identification

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

athi

-0.68

ovie

-0.66

 Carbuncle

-0.66

udeb

-0.64

jl

-0.64

 Scarlet

-0.63

FactoryReloaded

-0.62

 chanted

-0.59

 Cato

-0.58

BSD

-0.57

POSITIVE LOGITS

how

0.95

 whether

0.95

 whats

0.88

uate

0.83

whether

0.77

 exactly

0.75

 beforehand

0.74

 approximate

0.73

worthiness

0.72

 whereabouts

0.72

Activations Density 0.284%

keywords related to knowledge, identification, determination, and inference

phrases related to the processes of knowledge verification and identification

No Comments

No Known Activations