INDEX

Explanations

instances where something is being treated or perceived differently

oai_token-act-pair · gpt-3.5-turbo

situations or concepts that involve contrasting treatment or approaches

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Encyclopedia

-0.75

 Zion

-0.69

 Dmitry

-0.67

á

-0.66

Dou

-0.65

 Plate

-0.65

-0.64

 Upton

-0.63

advertising

-0.63

 Fritz

-0.62

POSITIVE LOGITS

iating

1.03

etheless

0.98

wcs

0.97

iates

0.94

 differently

0.93

 differentiated

0.89

iator

0.86

styles

0.85

different

0.84

 minded

0.83

Activations Density 0.005%

instances where something is being treated or perceived differently

situations or concepts that involve contrasting treatment or approaches

No Comments

No Known Activations