INDEX

Explanations

expressions of deep emotion and introspection

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

escription

-0.81

 favourable

-0.80

 authorised

-0.78

¥ŀ

-0.77

eatures

-0.75

 intended

-0.74

 detailed

-0.74

 dubbed

-0.74

 hailed

-0.73

 authorization

-0.72

POSITIVE LOGITS

 Anyway

1.63

 Especially

1.52

 Thankfully

1.32

 Luckily

1.31

 Besides

1.30

 Lastly

1.29

 Therefore

1.28

 Regardless

1.28

 Unless

1.27

Anyway

1.25

Activations Density 0.370%

expressions of deep emotion and introspection

No Comments

No Known Activations