INDEX

Explanations

phrases related to agreement or approval

oai_token-act-pair · gpt-3.5-turbo

statements related to opinions or personal sentiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

¶ħ

-0.59

bryce

-0.52

ected

-0.50

blogspot

-0.50

brids

-0.48

 Âł

-0.47

 Previously

-0.47

selling

-0.46

olitan

-0.46

chool

-0.45

POSITIVE LOGITS

 depends

0.50

 increments

0.48

 oneself

0.47

 behav

0.46

ain

0.44

 Almighty

0.43

 explan

0.43

TTC

0.43

 finish

0.42

 digest

0.42

Activations Density 2.003%

phrases related to agreement or approval

statements related to opinions or personal sentiments

No Comments

No Known Activations