INDEX

Explanations

words related to the working class

oai_token-act-pair · gpt-3.5-turbo

references to the working class and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

EStream

-0.79

hemy

-0.79

Ĥª

-0.78

asus

-0.73

netflix

-0.71

Ń·

-0.70

choes

-0.68

laus

-0.68

noxious

-0.67

antha

-0.67

POSITIVE LOGITS

 class

0.83

class

0.82

 stopp

0.82

horse

0.81

 conditions

0.79

 ethic

0.78

paces

0.75

classes

0.73

aday

0.72

hour

0.70

Activations Density 0.034%

words related to the working class

references to the working class and related concepts

No Comments

No Known Activations