INDEX

Explanations

phrases related to large groups or quantities happening at the same time

oai_token-act-pair · gpt-3.5-turbo

references to large groups or masses of people or entities

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

AUT

-0.71

ndra

-0.66

Own

-0.66

 Mobility

-0.66

 Ethnic

-0.63

KNOWN

-0.62

Balt

-0.62

 Niet

-0.62

 Promotion

-0.61

prime

-0.61

POSITIVE LOGITS

enance

0.85

 coating

0.81

ographed

0.78

uration

0.76

 paste

0.74

rette

0.74

hire

0.73

owder

0.71

 coated

0.71

 spraying

0.70

Activations Density 0.143%

phrases related to large groups or quantities happening at the same time

references to large groups or masses of people or entities

No Comments

No Known Activations