INDEX

Explanations

semiconductor-related terms, specifically related to the Zen architecture

oai_token-act-pair · gpt-3.5-turbo

references to Zen, likely in a technological or cultural context

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

riers

-0.79

âĢ¢âĢ¢âĢ¢âĢ¢

-0.77

 Lauder

-0.77

Ø©

-0.67

 Tire

-0.66

 Tolkien

-0.66

å¥³

-0.65

 Bagg

-0.65

 mileage

-0.64

 Scots

-0.63

POSITIVE LOGITS

eca

1.05

obia

0.96

zen

0.93

ith

0.87

emy

0.86

ical

0.84

pei

0.83

jet

0.83

itized

0.82

geist

0.82

Activations Density 0.015%

semiconductor-related terms, specifically related to the Zen architecture

references to Zen, likely in a technological or cultural context

No Comments

No Known Activations