INDEX

Explanations

information related to bridges

oai_token-act-pair · gpt-3.5-turbo

references to various "bridges" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

rub

-0.71

cub

-0.70

 recip

-0.68

otal

-0.66

 vacc

-0.64

Magazine

-0.64

acters

-0.63

 latex

-0.62

 giveaways

-0.62

 narc

-0.62

POSITIVE LOGITS

 Bridge

4.06

Bridge

2.67

 bridge

2.42

 Bridges

2.17

 bridges

2.01

bridge

1.93

 Crossing

1.21

 Tunnel

1.20

 River

1.19

 Boat

1.17

Activations Density 0.012%

information related to bridges

references to various "bridges" in different contexts

No Comments

No Known Activations