Class: LlamaDeuce
Llama2 LLM implementation
Hierarchy
BaseLLM↳
LlamaDeuce
Constructors
constructor
• new LlamaDeuce(init?)
Parameters
| Name | Type |
|---|---|
init? | Partial<LlamaDeuce> |
Overrides
BaseLLM.constructor
Defined in
packages/core/src/llm/LLM.ts:358
Properties
chatStrategy
• chatStrategy: DeuceChatStrategy
Defined in
packages/core/src/llm/LLM.ts:352
maxTokens
• Optional maxTokens: number
Defined in
packages/core/src/llm/LLM.ts:355
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:351
replicateSession
• replicateSession: ReplicateSession
Defined in
packages/core/src/llm/LLM.ts:356
temperature
• temperature: number
Defined in
packages/core/src/llm/LLM.ts:353
topP
• topP: number
Defined in
packages/core/src/llm/LLM.ts:354
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:378
Methods
chat
▸ chat(params): Promise<AsyncIterable<ChatResponseChunk>>
Parameters
| Name | Type |
|---|---|
params | LLMChatParamsStreaming |
Returns
Promise<AsyncIterable<ChatResponseChunk>>
Overrides
BaseLLM.chat
Defined in
packages/core/src/llm/LLM.ts:516
▸ chat(params): Promise<ChatResponse>
Parameters
| Name | Type |
|---|---|
params | LLMChatParamsNonStreaming |
Returns
Promise<ChatResponse>
Overrides
BaseLLM.chat
Defined in
packages/core/src/llm/LLM.ts:519
complete
▸ complete(params): Promise<AsyncIterable<CompletionResponse>>
Parameters
| Name | Type |
|---|---|
params | LLMCompletionParamsStreaming |
Returns
Promise<AsyncIterable<CompletionResponse>>
Inherited from
BaseLLM.complete
Defined in
packages/core/src/llm/base.ts:18
▸ complete(params): Promise<CompletionResponse>
Parameters
| Name | Type |
|---|---|
params | LLMCompletionParamsNonStreaming |
Returns
Promise<CompletionResponse>
Inherited from
BaseLLM.complete
Defined in
packages/core/src/llm/base.ts:21
mapMessageTypeA16Z
▸ mapMessageTypeA16Z(messageType): string
Parameters
| Name | Type |
|---|---|
messageType | MessageType |
Returns
string
Defined in
packages/core/src/llm/LLM.ts:425
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:389
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:411
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:438
tokens
▸ tokens(messages): number
Parameters
| Name | Type |
|---|---|
messages | ChatMessage[] |
Returns
number
Overrides
BaseLLM.tokens