INDEX

Explanations

numerical intervals or measurements

oai_token-act-pair · gpt-3.5-turbo

references to intervals and their variations

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

son

-0.77

ongyang

-0.75

tty

-0.75

HCR

-0.72

kie

-0.72

ryn

-0.71

eger

-0.67

iard

-0.66

osis

-0.66

gewater

-0.66

POSITIVE LOGITS

 intervals

1.26

 interval

1.18

 periods

0.75

ftime

0.71

arella

0.71

 mund

0.70

 gratification

0.70

 occas

0.69

 ende

0.68

 between

0.68

Activations Density 0.010%

numerical intervals or measurements

references to intervals and their variations

No Comments

No Known Activations