INDEX

Explanations

terms related to news syndication and broadcasting

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ghazi

-0.76

 Ness

-0.70

FUL

-0.68

 slideshow

-0.63

 Miko

-0.63

 pity

-0.62

++++++++++++++++

-0.61

 largeDownload

-0.60

 uphill

-0.60

 Germ

-0.59

POSITIVE LOGITS

icate

1.54

ication

1.45

icated

1.43

icates

1.42

icator

1.32

icating

1.25

ications

1.15

icators

1.13

emic

1.06

ical

1.06

Activations Density 0.019%

terms related to news syndication and broadcasting

No Comments

No Known Activations