INDEX

Explanations

information related to historical figures, especially Greek and Roman philosophers and historians

oai_token-act-pair · gpt-3.5-turbo

references to historical or philosophical texts and figures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hower

-0.86

Pak

-0.83

 motel

-0.81

 UNCLASSIFIED

-0.80

ipeg

-0.79

ski

-0.78

raviolet

-0.78

olson

-0.76

 Blackhawks

-0.75

 runway

-0.74

POSITIVE LOGITS

 Romans

1.65

 Athen

1.49

 Corinth

1.49

 Greeks

1.44

 Herod

1.43

 Greek

1.40

 Roman

1.39

Byz

1.37

 Carth

1.35

 Plato

1.34

Activations Density 0.319%

information related to historical figures, especially Greek and Roman philosophers and historians

references to historical or philosophical texts and figures

No Comments

No Known Activations