INDEX

Explanations

references to specific locations, with a focus on Kenya and Tanzania

oai_token-act-pair · gpt-3.5-turbo

references to Kenya and related geographical contexts

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

alty

-0.91

orsche

-0.82

 accur

-0.77

ÑĮ

-0.75

Ö¼

-0.74

ellation

-0.73

 predictive

-0.72

enegger

-0.71

mble

-0.70

ALSE

-0.70

POSITIVE LOGITS

 Kenya

0.98

 Tanzania

0.91

 Haram

0.86

istani

0.85

 Ethiopia

0.85

 Leone

0.80

istan

0.78

 Kenyan

0.77

ji

0.77

 Airways

0.77

Activations Density 0.007%

references to specific locations, with a focus on Kenya and Tanzania

references to Kenya and related geographical contexts

No Comments

No Known Activations