INDEX

Explanations

locations or regions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating geographic locations and their relationships

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

-0.77

tick

-0.73

%%

-0.70

 nostalg

-0.70

 ALWAYS

-0.70

UID

-0.68

ENSE

-0.64

orius

-0.63

 VIDEOS

-0.63

ÃįÃį

-0.62

POSITIVE LOGITS

 neighbouring

1.19

 elsewhere

1.17

 neighboring

1.16

 other

1.09

 others

1.04

 adjoining

1.01

 mainland

0.86

 another

0.83

 Other

0.82

 surrounding

0.80

Activations Density 0.358%

locations or regions

phrases indicating geographic locations and their relationships

No Comments

No Known Activations