INDEX

Explanations

names of people

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

shire

-0.86

iaries

-0.72

regor

-0.68

 uncanny

-0.64

Ö¼

-0.61

ionage

-0.61

 adoptive

-0.61

line

-0.57

tein

-0.56

 extrap

-0.56

POSITIVE LOGITS

Ã±a

1.20

Ã±

1.13

uthor

1.12

Ã±o

1.03

qual

1.02

ppa

0.99

zzo

0.98

heed

0.96

qa

0.95

qua

0.95

Activations Density 0.109%

names of people

No Comments

No Known Activations