INDEX

Explanations

information related to people's professions and roles

oai_token-act-pair · gpt-3.5-turbo

occupational titles and roles

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

yles

-0.85

ouses

-0.85

days

-0.81

éĹĺ

-0.80

akings

-0.79

cific

-0.78

mates

-0.76

adas

-0.75

onies

-0.74

ONSORED

-0.74

POSITIVE LOGITS

 educator

1.50

 entrepreneur

1.45

 musician

1.44

 lecturer

1.42

 poet

1.39

 novelist

1.39

 philosopher

1.37

 filmmaker

1.37

 broadcaster

1.35

 writer

1.34

Activations Density 0.143%

information related to people's professions and roles

occupational titles and roles

No Comments

No Known Activations