INDEX

Explanations

topics or subjects being discussed or focused on in text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "topic."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Zimmer

-0.70

igned

-0.70

ignt

-0.69

hovah

-0.66

inyl

-0.65

leeve

-0.65

ardo

-0.64

ramid

-0.63

omp

-0.62

ATES

-0.61

POSITIVE LOGITS

 topics

0.95

Topic

0.92

matter

0.91

 topic

0.88

topic

0.87

 Topics

0.83

TextColor

0.80

Topics

0.78

userc

0.78

forum

0.75

Activations Density 0.022%

topics or subjects being discussed or focused on in text

occurrences of the word "topic."

No Comments

No Known Activations