INDEX

Explanations

numeric data and statistics related to viewership and ratings of television programs

oai_token-act-pair · gpt-3.5-turbo

numeric and statistical data related to viewership and ratings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

bard

-0.82

Alz

-0.78

 Hast

-0.77

 Wass

-0.72

 contra

-0.71

 healer

-0.71

 Forth

-0.71

ulic

-0.70

 laundering

-0.69

 Sass

-0.69

POSITIVE LOGITS

 viewership

1.67

 Nielsen

1.44

 viewers

1.44

 ratings

1.41

 Ratings

1.20

 rating

1.20

 viewer

1.17

 airing

1.16

NBC

1.13

 audience

1.13

Activations Density 0.137%

numeric data and statistics related to viewership and ratings of television programs

numeric and statistical data related to viewership and ratings

No Comments

No Known Activations