INDEX

Explanations

official titles or positions within a formal setting such as government

oai_token-act-pair · gpt-3.5-turbo

references to government officials and their roles or statements

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

Population

-0.75

bows

-0.69

pmwiki

-0.68

 captcha

-0.67

Mods

-0.65

 landfall

-0.65

inventoryQuantity

-0.64

DragonMagazine

-0.63

 losers

-0.62

 killers

-0.62

POSITIVE LOGITS

 briefed

1.36

 familiar

1.21

 told

1.11

 quoted

1.06

 cautioned

1.06

who

1.06

 interviewed

1.01

 testified

1.00

 cited

0.99

amiliar

0.98

Activations Density 0.073%

official titles or positions within a formal setting such as government

references to government officials and their roles or statements

No Comments

No Known Activations