INDEX

Explanations

references to specific entities or proper nouns in the context of announcements, events, and news

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

INESS

-0.73

ELS

-0.68

DERR

-0.67

hesive

-0.66

Plug

-0.63

ITNESS

-0.62

OTAL

-0.61

Sus

-0.60

esville

-0.60

scl

-0.59

POSITIVE LOGITS

pheus

1.18

phans

1.13

thodox

1.11

lando

1.01

leans

1.01

chid

1.00

ussia

0.99

izons

0.98

bid

0.95

rent

0.94

Activations Density 0.808%

references to specific entities or proper nouns in the context of announcements, events, and news

No Comments

No Known Activations