INDEX

Explanations

negative consequences or criticisms associated with government actions or policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

xual

-0.92

 casc

-0.77

 destro

-0.76

 altern

-0.74

 regener

-0.72

 exha

-0.70

 racing

-0.69

 ende

-0.68

 transitions

-0.68

 seiz

-0.68

POSITIVE LOGITS

WASHINGTON

1.51

Posted

1.39

Published

1.37

Updated

1.28

SHARE

1.26

Author

1.23

Former

1.23

LOS

1.23

SAN

1.22

Washington

1.21

Activations Density 0.294%

negative consequences or criticisms associated with government actions or policies

No Comments

No Known Activations