INDEX

Explanations

references to the country Canada

oai_token-act-pair · gpt-3.5-turbo

mentions of Canada

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

urations

-0.94

bler

-0.80

ãĥł

-0.78

inse

-0.73

ysis

-0.71

utm

-0.71

udeb

-0.71

tle

-0.71

vel

-0.70

uers

-0.70

POSITIVE LOGITS

 Canada

1.19

 Canadians

1.17

 Scotia

1.12

Canada

1.06

 Canad

1.00

 Goose

0.95

 Centre

0.92

 Canadian

0.88

 Forces

0.88

 Trudeau

0.88

Activations Density 0.016%

references to the country Canada

mentions of Canada

No Comments

No Known Activations