INDEX

Explanations

phrases related to situations or actions happening in front of someone

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

zn

-0.68

avorite

-0.68

cci

-0.67

chini

-0.67

redited

-0.66

 Strait

-0.64

zie

-0.63

 Preferred

-0.60

ajor

-0.60

risome

-0.59

POSITIVE LOGITS

aday

0.89

of

0.81

matter

0.75

 thereof

0.74

runners

0.74

isp

0.74

woods

0.71

iers

0.71

of

0.70

screen

0.69

Activations Density 0.013%

phrases related to situations or actions happening in front of someone

No Comments

No Known Activations