INDEX

Explanations

words related to names or titles

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific place or notable reference within the context of the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ELL

-0.75

 Occupations

-0.73

 Worker

-0.71

EED

-0.70

SUP

-0.67

IELD

-0.66

IRED

-0.65

 Hurricanes

-0.64

pocket

-0.64

 Healer

-0.63

POSITIVE LOGITS

Dol

0.97

umin

0.94

leys

0.90

isco

0.88

ores

0.88

icates

0.87

imar

0.86

icate

0.86

hyde

0.85

onna

0.83

Activations Density 0.018%

words related to names or titles

mentions of a specific place or notable reference within the context of the text

No Comments

No Known Activations