INDEX

Explanations

content related to political and social issues with a focus on topics like medical procedures, cultural norms, opinions on beliefs, and government actions

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

contact

-0.80

estones

-0.78

 sighting

-0.77

 completion

-0.74

 Finder

-0.74

pleted

-0.73

 gauge

-0.71

availability

-0.70

hooting

-0.70

oother

-0.70

POSITIVE LOGITS

 hypocrisy

1.63

 hypocritical

1.51

 disingen

1.41

 arrogance

1.37

 coward

1.37

 misguided

1.34

 irresponsible

1.34

 hypoc

1.33

 bigotry

1.32

 hypocr

1.32

Activations Density 9.722%

content related to political and social issues with a focus on topics like medical procedures, cultural norms, opinions on beliefs, and government actions

No Comments

No Known Activations