INDEX

Explanations

references to the location "Grand"

oai_token-act-pair · gpt-3.5-turbo

references to the word "Grand."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

-0.98

tics

-0.81

ijk

-0.69

 stocks

-0.68

ylon

-0.67

-0.66

ayers

-0.65

irming

-0.64

lihood

-0.64

utics

-0.64

POSITIVE LOGITS

 Prix

1.20

father

1.03

mother

0.98

master

0.94

pa

0.89

 Theft

0.88

iosity

0.87

 Grand

0.80

view

0.80

parents

0.79

Activations Density 0.007%

references to the location "Grand"

references to the word "Grand."

No Comments

No Known Activations