INDEX

Explanations

words related to government policies, laws, and regulations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oris

-0.55

Nar

-0.54

 glim

-0.53

 Phar

-0.53

 Hitch

-0.52

IRC

-0.52

 Fulton

-0.52

VB

-0.51

ourning

-0.51

 sophomore

-0.51

POSITIVE LOGITS

 regardless

1.09

 irrespective

1.07

 differently

1.06

 whereas

1.02

 instead

1.00

 anyway

0.99

 accordingly

0.98

 uncond

0.97

 rather

0.95

 wherever

0.92

Activations Density 0.838%

words related to government policies, laws, and regulations

No Comments

No Known Activations