INDEX

Explanations

references to Canadian entities or elements

oai_token-act-pair · gpt-3.5-turbo

mentions of Canada or Canadian entities

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

bler

-0.87

mble

-0.80

inse

-0.79

hots

-0.79

uer

-0.76

itzer

-0.75

blers

-0.75

oldemort

-0.72

%]

-0.72

atz

-0.72

POSITIVE LOGITS

 Forces

1.21

 Tire

1.01

 Centre

0.92

apolis

0.90

 provinces

0.89

 Armed

0.89

 Canadians

0.88

 Institutes

0.87

 Goose

0.85

 Rockies

0.85

Activations Density 0.023%

references to Canadian entities or elements

mentions of Canada or Canadian entities

No Comments

No Known Activations