INDEX

Explanations

references to specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

references to locations and their associated cultural or political contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Wisconsin

-1.08

 Milwaukee

-1.08

Michigan

-1.03

 Pittsburgh

-0.97

 Wisconsin

-0.91

 Rutherford

-0.90

Pitt

-0.88

 Steelers

-0.87

Madison

-0.86

 Penguins

-0.86

POSITIVE LOGITS

 Thai

2.14

 Myanmar

1.99

 Thailand

1.97

 Bangkok

1.96

 Rohingya

1.85

 Indonesian

1.84

 Burma

1.82

 Cambod

1.81

 Jakarta

1.77

 Cambodia

1.71

Activations Density 0.464%

references to specific geographical locations

references to locations and their associated cultural or political contexts

No Comments

No Known Activations