INDEX

Explanations

terms related to categories and classifications, such as names, places, religions, histories, and ecology

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Operation

-0.72

asus

-0.68

atorium

-0.62

culosis

-0.62

ilogy

-0.62

noon

-0.61

 Patriarch

-0.59

iak

-0.58

 apostle

-0.58

Bed

-0.58

POSITIVE LOGITS

paces

1.16

etc

1.15

pace

1.13

peed

1.12

mith

1.11

afety

1.06

etting

1.02

hips

0.96

gal

0.96

 abound

0.94

Activations Density 0.481%

terms related to categories and classifications, such as names, places, religions, histories, and ecology

No Comments

No Known Activations