INDEX

Explanations

references to the country Yugoslavia

oai_token-act-pair · gpt-3.5-turbo

references to Yugoslavia and its associated nationalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WARD

-0.85

ORGE

-0.78

reads

-0.75

oyer

-0.70

ellation

-0.69

enting

-0.69

encers

-0.67

 gust

-0.67

 glim

-0.66

Score

-0.66

POSITIVE LOGITS

 Alban

1.09

 Serbia

1.03

Äį

1.01

 Croatia

0.96

 Bosnia

0.93

 Ukrain

0.93

oslov

0.92

Äĩ

0.90

 Herz

0.90

oslav

0.87

Activations Density 0.030%

references to the country Yugoslavia

references to Yugoslavia and its associated nationalities

No Comments

No Known Activations