INDEX

Explanations

references to names, possibly related to characters or individuals

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

espie

-0.80

undy

-0.73

agles

-0.73

elaide

-0.72

tering

-0.72

riers

-0.72

 Lyons

-0.70

ftime

-0.70

Colorado

-0.70

 Bland

-0.69

POSITIVE LOGITS

ichi

1.41

yen

1.31

 Yosh

1.31

oka

1.30

atsu

1.26

Yen

1.26

 Mats

1.25

etsu

1.20

 Tsuk

1.19

jin

1.17

Activations Density 1.905%

references to names, possibly related to characters or individuals

No Comments

No Known Activations