INDEX

Explanations

phrases related to news or updates

oai_token-act-pair · gpt-3.5-turbo

instances of advertisements or promotional content within a text

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

 homebrew

-0.71

 offshore

-0.63

addin

-0.63

 honoured

-0.59

mbuds

-0.58

 estranged

-0.57

 unaccount

-0.57

ktop

-0.56

 paed

-0.56

 travelling

-0.56

POSITIVE LOGITS

SPONSORED

0.91

ccording

0.83

Related

0.80

Copyright

0.79

JUST

0.79

VERTISEMENT

0.77

ATT

0.74

Support

0.72

Anonymous

0.72

Advertisement

0.71

Activations Density 0.054%

phrases related to news or updates

instances of advertisements or promotional content within a text

No Comments

No Known Activations