Class: LlamaDeuce
Llama2 LLM implementation
Hierarchy
↳
LlamaDeuce
Constructors
constructor
• new LlamaDeuce(init?
)
Parameters
Name | Type |
---|---|
init? | Partial <LlamaDeuce > |
Overrides
Defined in
packages/core/src/llm/LLM.ts:487
Properties
chatStrategy
• chatStrategy: DeuceChatStrategy
Defined in
packages/core/src/llm/LLM.ts:481
maxTokens
• Optional
maxTokens: number
Defined in
packages/core/src/llm/LLM.ts:484
model
• model: "Llama-2-70b-chat-old"
| "Llama-2-70b-chat-4bit"
| "Llama-2-13b-chat-old"
| "Llama-2-13b-chat-4bit"
| "Llama-2-7b-chat-old"
| "Llama-2-7b-chat-4bit"
Defined in
packages/core/src/llm/LLM.ts:480
replicateSession
• replicateSession: ReplicateSession
Defined in
packages/core/src/llm/LLM.ts:485
temperature
• temperature: number
Defined in
packages/core/src/llm/LLM.ts:482
topP
• topP: number
Defined in
packages/core/src/llm/LLM.ts:483
Accessors
metadata
• get
metadata(): Object
Returns
Object
Name | Type |
---|---|
contextWindow | number |
maxTokens | undefined | number |
model | "Llama-2-70b-chat-old" | "Llama-2-70b-chat-4bit" | "Llama-2-13b-chat-old" | "Llama-2-13b-chat-4bit" | "Llama-2-7b-chat-old" | "Llama-2-7b-chat-4bit" |
temperature | number |
tokenizer | undefined |
topP | number |
Overrides
BaseLLM.metadata
Defined in
packages/core/src/llm/LLM.ts:507
Methods
chat
▸ chat(params
): Promise
<AsyncIterable
<ChatResponseChunk
>>
Get a chat response from the LLM
Parameters
Name | Type |
---|---|
params | LLMChatParamsStreaming |
Returns
Promise
<AsyncIterable
<ChatResponseChunk
>>
Overrides
Defined in
packages/core/src/llm/LLM.ts:645
▸ chat(params
): Promise
<ChatResponse
>
Parameters
Name | Type |
---|---|
params | LLMChatParamsNonStreaming |
Returns
Promise
<ChatResponse
>
Overrides
Defined in
packages/core/src/llm/LLM.ts:648
complete
▸ complete(params
): Promise
<AsyncIterable
<CompletionResponse
>>
Get a prompt completion from the LLM
Parameters
Name | Type |
---|---|
params | LLMCompletionParamsStreaming |
Returns
Promise
<AsyncIterable
<CompletionResponse
>>
Inherited from
Defined in
packages/core/src/llm/LLM.ts:138
▸ complete(params
): Promise
<CompletionResponse
>
Parameters
Name | Type |
---|---|
params | LLMCompletionParamsNonStreaming |
Returns
Promise
<CompletionResponse
>
Inherited from
Defined in
packages/core/src/llm/LLM.ts:141
mapMessageTypeA16Z
▸ mapMessageTypeA16Z(messageType
): string
Parameters
Name | Type |
---|---|
messageType | MessageType |
Returns
string
Defined in
packages/core/src/llm/LLM.ts:554
mapMessagesToPrompt
▸ mapMessagesToPrompt(messages
): Object
Parameters
Name | Type |
---|---|
messages | ChatMessage [] |
Returns
Object
Name | Type |
---|---|
prompt | string |
systemPrompt | any |
Defined in
packages/core/src/llm/LLM.ts:518
mapMessagesToPromptA16Z
▸ mapMessagesToPromptA16Z(messages
): Object
Parameters
Name | Type |
---|---|
messages | ChatMessage [] |
Returns
Object
Name | Type |
---|---|
prompt | string |
systemPrompt | undefined |
Defined in
packages/core/src/llm/LLM.ts:540
mapMessagesToPromptMeta
▸ mapMessagesToPromptMeta(messages
, opts?
): Object
Parameters
Name | Type |
---|---|
messages | ChatMessage [] |
opts? | Object |
opts.replicate4Bit? | boolean |
opts.withBos? | boolean |
opts.withNewlines? | boolean |
Returns
Object
Name | Type |
---|---|
prompt | string |
systemPrompt | any |
Defined in
packages/core/src/llm/LLM.ts:567
tokens
▸ tokens(messages
): number
Calculates the number of tokens needed for the given chat messages
Parameters
Name | Type |
---|---|
messages | ChatMessage [] |
Returns
number