INDEX

Explanations

proper nouns ending with a single letter 'E'

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Bleach

-0.73

 minds

-0.68

ifiable

-0.63

 Anthem

-0.59

Io

-0.58

 Hera

-0.58

 Painter

-0.58

 Akira

-0.58

 Vulkan

-0.57

ãĤ¤ãĥĪ

-0.57

POSITIVE LOGITS

ighty

1.24

LECT

1.21

tymology

1.21

lements

1.20

clipse

1.17

agle

1.15

TERN

1.14

cosystem

1.14

ASY

1.13

ighth

1.13

Activations Density 0.032%

proper nouns ending with a single letter 'E'

No Comments

No Known Activations