INDEX

Explanations

locations or places described in the text, particularly focusing on places within specific regions or countries

oai_token-act-pair · gpt-3.5-turbo

instances of the preposition "in" related to locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

WORK

-0.72

Reviewer

-0.69

 debuted

-0.68

Desk

-0.67

rooms

-0.65

 lodged

-0.65

trump

-0.64

 desks

-0.63

acters

-0.63

TPP

-0.63

POSITIVE LOGITS

 southwestern

1.40

 southwest

1.39

 northern

1.39

 southeastern

1.34

 northeastern

1.32

 northwest

1.32

 southern

1.32

 northeast

1.30

 northwestern

1.30

 southeast

1.26

Activations Density 0.243%

locations or places described in the text, particularly focusing on places within specific regions or countries

instances of the preposition "in" related to locations

No Comments

No Known Activations