INDEX

Explanations

instances of the word "goods" and "services"

oai_token-act-pair · gpt-3.5-turbo

references to goods and services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

adian

-0.75

cair

-0.75

ODE

-0.71

neum

-0.67

NING

-0.67

otos

-0.67

arios

-0.65

uberty

-0.63

ULAR

-0.63

ederal

-0.62

POSITIVE LOGITS

 goods

1.00

loads

0.78

gob

0.78

bags

0.74

maker

0.73

washer

0.73

makers

0.72

andise

0.72

 exports

0.72

meal

0.72

Activations Density 0.025%

instances of the word "goods" and "services"

references to goods and services

No Comments

No Known Activations