INDEX

Explanations

actions related to the act of shaving

oai_token-act-pair · gpt-3.5-turbo

terms related to shaving and haircuts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

alez

-0.81

encia

-0.76

 Guilty

-0.68

 Crom

-0.66

gom

-0.66

 Miranda

-0.64

ologue

-0.62

Cass

-0.62

phis

-0.62

iege

-0.61

POSITIVE LOGITS

 shaving

1.22

 shave

1.20

 shaved

1.05

utic

0.90

utical

0.83

utics

0.83

 scars

0.82

 blades

0.81

 trimmed

0.80

rette

0.78

Activations Density 0.007%

actions related to the act of shaving

terms related to shaving and haircuts

No Comments

No Known Activations