INDEX

Explanations

proper nouns related to lists

oai_token-act-pair · gpt-3.5-turbo

instances of the word "List" followed by various numbers

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

Downloadha

-0.85

icago

-0.79

rir

-0.72

artifacts

-0.71

irgin

-0.66

rity

-0.64

ulatory

-0.62

utherford

-0.61

 whist

-0.61

perty

-0.59

POSITIVE LOGITS

 List

1.20

ening

1.00

 Lists

0.98

erv

0.95

list

0.85

ener

0.85

witz

0.85

List

0.81

erves

0.81

ings

0.80

Activations Density 0.006%

proper nouns related to lists

instances of the word "List" followed by various numbers

No Comments

No Known Activations