INDEX

Explanations

keywords related to countries, particularly focusing on Hungary

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Hungary" 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

nesota

-0.87

phrine

-0.82

jad

-0.75

teen

-0.74

cloth

-0.74

iating

-0.68

-0.67

grades

-0.66

monds

-0.66

washed

-0.65

POSITIVE LOGITS

arians

0.98

 Budapest

0.87

 Hungarian

0.83

owitz

0.80

 Viktor

0.78

istani

0.77

Hung

0.77

ism

0.76

arian

0.74

 Hungary

0.73

Activations Density 0.120%

keywords related to countries, particularly focusing on Hungary

occurrences of the word "Hungary" and its variations

No Comments

No Known Activations