INDEX

Explanations

mentions of the word "watches" referring to timepieces or the act of observing

oai_token-act-pair · gpt-3.5-turbo

references to watches and related actions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 boost

-0.64

olution

-0.64

ax

-0.63

ASE

-0.59

 smear

-0.58

bru

-0.58

 hurdle

-0.58

olved

-0.58

¶

-0.57

POSITIVE LOGITS

 watches

3.72

 listens

1.63

atches

1.62

 dresses

1.50

 waits

1.47

 radios

1.46

 monitors

1.44

TVs

1.36

 watch

1.32

 watched

1.29

Activations Density 0.013%

mentions of the word "watches" referring to timepieces or the act of observing

references to watches and related actions

No Comments

No Known Activations