INDEX

Explanations

references to the concept of "home."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Home."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

terday

-0.77

"]=>

-0.77

ĸļ

-0.74

 McKenna

-0.71

OTUS

-0.68

utical

-0.68

 Twain

-0.68

 entirety

-0.67

uous

-0.67

 guiActiveUn

-0.66

POSITIVE LOGITS

 Home

1.09

Home

0.98

opath

0.87

home

0.86

coming

0.84

brew

0.84

 HOME

0.84

grown

0.83

opathy

0.82

owner

0.81

Activations Density 0.009%

references to the concept of "home."

occurrences of the word "Home."

No Comments

No Known Activations