INDEX

Explanations

entities or brand names mentioned in texts

oai_token-act-pair · gpt-3.5-turbo

references to streaming services and digital content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Rated

-0.95

instead

-0.79

Reply

-0.71

Atl

-0.71

çİĭ

-0.70

ITH

-0.67

KEN

-0.66

}.

-0.64

 ______

-0.64

OTH

-0.64

POSITIVE LOGITS

 caveats

0.72

 disclaimer

0.71

 arrests

0.66

 announcements

0.65

 nods

0.64

 occasional

0.64

 exceptions

0.64

 additions

0.64

 shuffle

0.63

 perks

0.62

Activations Density 0.316%

entities or brand names mentioned in texts

references to streaming services and digital content

No Comments

No Known Activations