INDEX

Explanations

locations within buildings

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" and references to specific locations or buildings

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

opsis

-0.77

arians

-0.73

alys

-0.72

lessness

-0.70

ogeneity

-0.70

fulness

-0.69

ohan

-0.69

eness

-0.68

eday

-0.68

agree

-0.67

POSITIVE LOGITS

 courthouse

1.13

 Statue

1.13

 renovated

1.10

 Museum

1.08

 Marriott

1.08

 Capitol

1.07

 Mansion

1.04

 Smithsonian

1.04

 Hotel

1.03

 Pavilion

1.03

Activations Density 0.277%

locations within buildings

occurrences of the word "the" and references to specific locations or buildings

No Comments

No Known Activations