INDEX

Explanations

information related to medical procedures and equipment

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ratulations

-0.76

reetings

-0.76

oho

-0.72

prise

-0.71

iversary

-0.70

pheus

-0.70

vana

-0.70

odiac

-0.70

holy

-0.70

Happy

-0.70

POSITIVE LOGITS

 inability

1.47

 unus

1.33

 inefficient

1.31

 unable

1.31

 Worse

1.29

 Unable

1.28

 inadequate

1.28

 deteriorated

1.28

 inadequ

1.25

 unreliable

1.24

Activations Density 3.948%

information related to medical procedures and equipment

No Comments

No Known Activations