Foundation Models on AWS Word Search
MediumFind each term hidden in the grid. Selecting a word reveals its definition and a link to study it in depth.
A
I
U
G
X
R
G
E
A
M
A
Z
O
N
Q
J
K
P
R
D
J
A
Y
A
V
V
Y
V
T
F
T
C
U
X
S
R
S
N
T
J
G
O
T
Y
Z
R
O
I
P
M
D
F
U
D
N
Z
P
P
B
J
A
R
O
Y
H
C
V
T
I
O
M
H
E
X
F
T
Y
V
T
Y
F
X
K
M
O
M
D
B
Y
W
S
T
R
H
N
E
N
L
R
W
R
N
V
L
B
P
R
V
S
T
U
G
P
Y
O
P
Q
E
S
P
M
A
G
N
H
Z
U
T
C
G
I
E
Z
S
P
U
P
O
C
F
X
N
K
H
O
M
T
Z
V
S
J
C
X
C
N
O
I
T
N
E
T
T
A
F
U
U
A
J
B
E
S
N
A
Y
J
R
O
G
K
T
C
Y
E
I
J
Z
E
W
G
P
O
X
N
Q
Q
G
W
G
B
Z
Z
X
Y
J
L
X
Z
F
L
R
K
D
T
E
R
P
R
E
T
N
I
K
B
Q
V
Drag across letters, or tap the first and last letter. On a keyboard, use arrows + Enter.
Terms in this set
- Prompt The input text or instruction given to a generative model.
- Chunking Splitting documents into smaller pieces so they can be embedded and retrieved.
- Attention The mechanism that lets a model weigh the relevance of different tokens to each other.
- Nondeterminism The property that a generative model can return different outputs for the same prompt.
- Interpretability The degree to which a human can understand how a model arrives at its output.
- Context Window The maximum number of tokens a model can consider in a single request.
- Bedrock Amazon Bedrock — a managed service offering foundation models through a single API.
- JumpStart SageMaker JumpStart — a hub of pre-built models and solution templates.
- PartyRock Amazon Bedrock PartyRock — a no-code playground for building generative AI apps.
- Amazon Q AWS's generative AI–powered assistant for business and developer tasks.