INDEX

Explanations

phrases related to products or services offered by The New York Times

oai_token-act-pair · gpt-3.5-turbo

mentions of "The New York Times" and its related elements

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

gow

-0.88

stood

-0.84

gart

-0.77

closed

-0.77

amiya

-0.75

lain

-0.74

verb

-0.74

gap

-0.70

anon

-0.70

Â¶

-0.67

POSITIVE LOGITS

 Newsletter

0.80

 editorial

0.79

 cartoons

0.78

 Presents

0.77

 sake

0.74

ournal

0.73

 weekly

0.72

 subscribers

0.71

 contributors

0.70

 affiliates

0.70

Activations Density 0.026%

phrases related to products or services offered by The New York Times

mentions of "The New York Times" and its related elements

No Comments

No Known Activations