INDEX

Explanations

references to the word "visa"

oai_token-act-pair · gpt-3.5-turbo

references to visas and immigration processes

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

ideshow

-0.72

ocrat

-0.68

worn

-0.67

idges

-0.66

sonian

-0.66

MpServer

-0.65

tes

-0.64

utenberg

-0.64

 Crom

-0.63

 Milky

-0.63

POSITIVE LOGITS

 visas

1.39

 visa

1.36

 versa

1.25

 waiver

0.82

 waivers

0.79

 holders

0.77

 passports

0.76

 lottery

0.75

 Visa

0.74

uana

0.74

Activations Density 0.008%

references to the word "visa"

references to visas and immigration processes

No Comments

No Known Activations