INDEX

Explanations

proper nouns, specifically names

oai_token-act-pair · gpt-3.5-turbo

mentions of locations or geographical identifiers

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

 plat

-0.68

 dece

-0.66

 wound

-0.63

 sheep

-0.62

 leve

-0.60

minster

-0.59

 calc

-0.59

 deeds

-0.58

 Hayward

-0.58

gem

-0.58

POSITIVE LOGITS

AN

3.90

ANS

2.53

ANI

2.22

ANA

2.05

ANE

2.01

ANN

1.94

ANT

1.86

ans

1.78

ANC

1.70

ANY

1.62

Activations Density 0.012%

proper nouns, specifically names

mentions of locations or geographical identifiers

No Comments

No Known Activations