INDEX

Explanations

mentions of news organizations and content

oai_token-act-pair · gpt-3.5-turbo

references to news-related content and alerts

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

igham

-0.79

amiya

-0.79

ierre

-0.76

ersen

-0.67

ngth

-0.66

onomous

-0.64

 exclusion

-0.64

ãĤ¶

-0.63

llan

-0.63

cel

-0.61

POSITIVE LOGITS

clinton

0.71

 eleph

0.68

 Thrones

0.65

 Newsp

0.65

yahoo

0.64

print

0.63

 Mubarak

0.63

zai

0.62

 fodder

0.61

TMZ

0.61

Activations Density 0.204%

mentions of news organizations and content

references to news-related content and alerts

No Comments

No Known Activations