Inference & Customization Word Search
MediumFind each term hidden in the grid. Selecting a word reveals its definition and a link to study it in depth.
T
Q
Q
G
A
F
I
N
E
T
U
N
E
B
B
O
K
N
O
W
L
E
D
G
E
Y
E
X
Y
D
P
D
I
N
C
O
N
T
E
X
T
A
G
U
T
K
D
I
M
A
X
T
O
K
E
N
S
Q
M
R
Q
E
Y
H
R
H
Y
I
B
W
Q
O
E
W
D
P
P
S
A
M
N
G
H
U
J
F
K
Y
R
V
B
R
G
S
S
B
M
C
K
M
N
U
D
D
R
D
E
T
E
M
P
E
R
A
T
U
R
E
O
R
R
T
R
M
O
F
M
A
E
C
B
E
T
Q
O
O
R
R
G
R
T
N
E
P
F
A
C
P
A
Y
T
A
Y
K
Q
A
S
S
H
T
E
U
D
I
P
C
I
K
C
C
C
L
N
F
V
O
O
J
S
E
E
N
H
O
Q
Q
M
E
G
F
T
P
F
H
W
V
S
A
J
J
N
T
P
M
U
Z
M
P
G
J
K
E
J
S
B
Q
I
O
Z
L
R
K
C
Z
I
Drag across letters, or tap the first and last letter. On a keyboard, use arrows + Enter.
Terms in this set
- Temperature An inference parameter controlling randomness — higher values make output more varied.
- Top-K A sampling setting that limits token choice to the K most likely options.
- Top-P Nucleus sampling — choosing from the smallest set of tokens whose probabilities sum to P.
- Max Tokens An inference parameter capping the length of the model's response.
- RAG Retrieval-Augmented Generation — grounding model answers in relevant retrieved documents.
- Vector Database A store of embeddings that supports fast similarity search for retrieval.
- Knowledge Base Amazon Bedrock Knowledge Bases — managed RAG that connects FMs to your data sources.
- OpenSearch Amazon OpenSearch Service — a search and analytics engine usable as a vector store for RAG.
- pgvector A PostgreSQL/Aurora extension that stores and searches embedding vectors.
- Fine-tuning Customizing a foundation model by further training it on labeled, task-specific data.
- In-Context Learning Guiding a model's behavior by including instructions or examples directly in the prompt.
- Continued Pre-training Further training an FM on large amounts of unlabeled domain data to build domain knowledge.