INDEX

Explanations

terms related to North America

oai_token-act-pair · gpt-3.5-turbo

references to "North American" and its variations

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

raph

-0.91

mercial

-0.77

bread

-0.76

gerald

-0.75

itial

-0.75

enegger

-0.74

yard

-0.73

bern

-0.72

ernaut

-0.71

Mn

-0.71

POSITIVE LOGITS

 Distribution

0.80

 continent

0.78

 Languages

0.73

 Treaty

0.72

 Networks

0.71

 Wrestling

0.70

 Federation

0.67

 Hemisphere

0.66

 Telescope

0.66

isd

0.65

Activations Density 0.051%

terms related to North America

references to "North American" and its variations

No Comments

No Known Activations