INDEX

Explanations

references to someone's possession or ownership, typically indicated by the word "Your" followed by a noun

oai_token-act-pair · gpt-3.5-turbo

the word "Your" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

itud

-0.72

ilts

-0.70

vati

-0.70

aneers

-0.69

hess

-0.68

Æ

-0.68

ice

-0.68

Pac

-0.66

oldemort

-0.66

]'

-0.64

POSITIVE LOGITS

own

1.17

 mileage

1.12

 favorite

1.08

 favourite

1.05

ths

0.98

selves

0.91

 imagination

0.89

 choice

0.89

 fingertips

0.88

 browser

0.87

Activations Density 0.105%

references to someone's possession or ownership, typically indicated by the word "Your" followed by a noun

the word "Your" in various contexts

No Comments

No Known Activations