INDEX

Explanations

numeric values such as amounts, quantities, and percentages, as well as specific objects like library cards and a physical condition like HIV

oai_token-act-pair · gpt-3.5-turbo

phrases related to financial contributions and transactions

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

 Vaugh

-0.80

 disadvant

-0.79

_.

-0.74

 undermin

-0.74

 destro

-0.71

rul

-0.69

enegger

-0.68

 Seym

-0.63

ilib

-0.62

 challeng

-0.62

POSITIVE LOGITS

uties

0.54

 âĢº

0.53

mounted

0.48

 Hunts

0.43

HS

0.43

 Farming

0.43

LC

0.42

optional

0.42

united

0.42

chester

0.42

Activations Density 1.199%

numeric values such as amounts, quantities, and percentages, as well as specific objects like library cards and a physical condition like HIV

phrases related to financial contributions and transactions

No Comments

No Known Activations