INDEX

Explanations

references to drums

oai_token-act-pair · gpt-3.5-turbo

references to drums and drumming in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Agency

-0.76

 Tasman

-0.68

hiro

-0.67

 Mandela

-0.65

################

-0.65

 Territories

-0.64

Kiw

-0.64

ahime

-0.63

 Hitman

-0.62

 Australians

-0.60

POSITIVE LOGITS

mers

1.20

ming

1.05

sonian

1.02

beat

1.01

 drum

1.00

 drums

0.99

nas

0.97

sticks

0.96

staff

0.88

heads

0.87

Activations Density 0.010%

references to drums

references to drums and drumming in various contexts

No Comments

No Known Activations