INDEX

Explanations

mentions of "The New York Times."

oai_token-act-pair · gpt-3.5-turbo

references to "The" followed by news-related titles or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

llan

-0.78

uphem

-0.66

anwhile

-0.66

ilion

-0.65

pil

-0.64

CVE

-0.63

lain

-0.62

 loud

-0.62

erves

-0.61

hang

-0.60

POSITIVE LOGITS

 Verge

0.87

 Huffington

0.81

 Bulletin

0.78

New

0.75

odor

0.74

 Economist

0.72

New

0.72

 Hive

0.71

 Intercept

0.71

Chel

0.71

Activations Density 0.018%

mentions of "The New York Times."

references to "The" followed by news-related titles or organizations

No Comments

No Known Activations