INDEX

Explanations

specific references to historical figures and texts

oai_token-act-pair · gpt-3.5-turbo

references to historical figures and philosophical texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

hower

-0.99

Pak

-0.92

 Olson

-0.85

olson

-0.83

ipeg

-0.80

Wisconsin

-0.79

schild

-0.77

Canada

-0.77

 Pakistani

-0.76

 Tammy

-0.76

POSITIVE LOGITS

 Corinth

1.52

 Athen

1.49

 Pompe

1.47

 Romans

1.43

 Athens

1.38

 Carth

1.38

 Socrates

1.35

 Nero

1.35

 Lucius

1.33

 Maced

1.32

Activations Density 0.269%

specific references to historical figures and texts

references to historical figures and philosophical texts

No Comments

No Known Activations