INDEX

Explanations

mentions of locations where people live

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "live in."

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

casters

-0.84

catentry

-0.78

killed

-0.78

offer

-0.77

Ĳ

-0.75

¿½

-0.73

 culprit

-0.70

STAR

-0.69

 undone

-0.67

acter

-0.65

POSITIVE LOGITS

 accordance

1.13

ordinate

1.02

animate

0.93

 lieu

0.89

 vitro

0.88

 spite

0.87

 Denmark

0.85

 France

0.85

 vain

0.83

 exile

0.80

Activations Density 0.141%

mentions of locations where people live

occurrences of the phrase "live in."

No Comments

No Known Activations