INDEX

Explanations

highly opinionated or evaluative statements

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and their frequencies in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 requested

-0.83

 manually

-0.76

ascus

-0.75

 scheduled

-0.73

 comm

-0.71

 tracker

-0.71

 pending

-0.71

 jointly

-0.71

 discontinued

-0.71

lings

-0.70

POSITIVE LOGITS

 Hence

1.41

 Whereas

1.37

 Thus

1.33

 Consequently

1.24

 Whether

1.24

 Whatever

1.23

 Moreover

1.21

 Accordingly

1.21

 Likewise

1.20

 Indeed

1.20

Activations Density 0.488%

highly opinionated or evaluative statements

punctuation marks and their frequencies in the text

No Comments

No Known Activations