INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper nouns and significant names

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eco

-0.88

 Borg

-0.83

aldo

-0.81

Äĩ

-0.81

cyclopedia

-0.80

-0.78

-0.76

 Diego

-0.76

 Codex

-0.75

dos

-0.75

POSITIVE LOGITS

SH

1.55

SH

1.52

Sh

1.49

sh

1.47

 Shaw

1.39

ash

1.20

sh

1.20

Sh

1.20

 Shirley

1.20

shaw

1.18

Activations Density 0.314%

proper nouns

proper nouns and significant names

No Comments

No Known Activations