INDEX

Explanations

names of actors and characters from TV series and movies

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

 disag

-0.87

 centralized

-0.82

ulic

-0.81

 consolid

-0.80

Statistics

-0.80

 consolidate

-0.79

ritis

-0.74

 administr

-0.74

ickr

-0.74

osate

-0.74

POSITIVE LOGITS

 actress

1.42

 actor

1.42

 films

1.36

 movies

1.36

 actors

1.36

movie

1.35

 movie

1.31

 sitcom

1.27

 actresses

1.27

 Movie

1.26

Activations Density 2.524%

names of actors and characters from TV series and movies

No Comments

No Known Activations