Archives
/
langchain
mirror of https://github.com/hwchase17/langchain
2
0
Fork 1
Code Issues Packages Projects Releases Wiki Activity

EXPERIMENTAL Generic LLM wrapper to support chat model interface with configurable chat prompt format (#8295)

Browse Source 
## Update 2023-09-08

This PR now supports further models in addition to Lllama-2 chat models.
See [this comment](#issuecomment-1668988543) for further details. The
title of this PR has been updated accordingly.

## Original PR description

This PR adds a generic `Llama2Chat` model, a wrapper for LLMs able to
serve Llama-2 chat models (like `LlamaCPP`,
`HuggingFaceTextGenInference`, ...). It implements `BaseChatModel`,
converts a list of chat messages into the [required Llama-2 chat prompt
format](https://huggingface.co/blog/llama2#how-to-prompt-llama-2) and
forwards the formatted prompt as `str` to the wrapped `LLM`. Usage
example:

```python
# uses a locally hosted Llama2 chat model
llm = HuggingFaceTextGenInference(
    inference_server_url="http://127.0.0.1:8080/",
    max_new_tokens=512,
    top_k=50,
    temperature=0.1,
    repetition_penalty=1.03,
)

# Wrap llm to support Llama2 chat prompt format.
# Resulting model is a chat model
model = Llama2Chat(llm=llm)

messages = [
    SystemMessage(content="You are a helpful assistant."),
    MessagesPlaceholder(variable_name="chat_history"),
    HumanMessagePromptTemplate.from_template("{text}"),
]

prompt = ChatPromptTemplate.from_messages(messages)
memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
chain = LLMChain(llm=model, prompt=prompt, memory=memory)

# use chat model in a conversation
# ...
```

Also part of this PR are tests and a demo notebook.

- Tag maintainer: @hwchase17
- Twitter handle: `@mrt1nz`

---------

Co-authored-by: Erick Friis <erick@langchain.dev>
pull/13549/head
Martin Krasser 11 months ago committed by GitHub
parent c56faa6ef1
commit 79ed66f870
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
9 changed files with 1153 additions and 5 deletions
Show Stats Download Patch File Download Diff File

729
docs/docs/integrations/chat/llama2_chat.ipynb
Unescape Escape View File

@ -0,0 +1,729 @@
{
"cells": [
{
"cell_type": "markdown",
"id": "90a1faf2",
"metadata": {},
"source": [
"# Llama-2 Chat\n",
"\n",
"This notebook shows how to augment Llama-2 `LLM`s with the `Llama2Chat` wrapper to support the [Llama-2 chat prompt format](https://huggingface.co/blog/llama2#how-to-prompt-llama-2). Several `LLM` implementations in LangChain can be used as interface to Llama-2 chat models. These include [HuggingFaceTextGenInference](https://python.langchain.com/docs/integrations/llms/huggingface_textgen_inference), [LlamaCpp](https://python.langchain.com/docs/use_cases/question_answering/how_to/local_retrieval_qa), [GPT4All](https://python.langchain.com/docs/integrations/llms/gpt4all), ..., to mention a few examples. \n",
"\n",
"`Llama2Chat` is a generic wrapper that implements `BaseChatModel` and can therefore be used in applications as [chat model](https://python.langchain.com/docs/modules/model_io/models/chat/). `Llama2Chat` converts a list of [chat messages](https://python.langchain.com/docs/modules/model_io/models/chat/#messages) into the [required chat prompt format](https://huggingface.co/blog/llama2#how-to-prompt-llama-2) and forwards the formatted prompt as `str` to the wrapped `LLM`."
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "36c03540",
"metadata": {},
"outputs": [],
"source": [
"from langchain.chains import LLMChain\n",
"from langchain.memory import ConversationBufferMemory\n",
"from langchain_experimental.chat_models import Llama2Chat"
]
},
{
"cell_type": "markdown",
"id": "5c76910f",
"metadata": {},
"source": [
"For the chat application examples below, we'll use the following chat `prompt_template`:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "9bbfaf3a",
"metadata": {},
"outputs": [],
"source": [
"from langchain.prompts.chat import (\n",
" ChatPromptTemplate,\n",
" HumanMessagePromptTemplate,\n",
" MessagesPlaceholder,\n",
")\n",
"from langchain.schema import SystemMessage\n",
"\n",
"template_messages = [\n",
" SystemMessage(content=\"You are a helpful assistant.\"),\n",
" MessagesPlaceholder(variable_name=\"chat_history\"),\n",
" HumanMessagePromptTemplate.from_template(\"{text}\"),\n",
"]\n",
"prompt_template = ChatPromptTemplate.from_messages(template_messages)"
]
},
{
"cell_type": "markdown",
"id": "2f3343b7",
"metadata": {},
"source": [
"## Chat with Llama-2 via `HuggingFaceTextGenInference` LLM"
]
},
{
"cell_type": "markdown",
"id": "2ff99380",
"metadata": {},
"source": [
"A [HuggingFaceTextGenInference](https://python.langchain.com/docs/integrations/llms/huggingface_textgen_inference) LLM encapsulates access to a [text-generation-inference](https://github.com/huggingface/text-generation-inference) server. In the following example, the inference server serves a [meta-llama/Llama-2-13b-chat-hf](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) model. It can be started locally with:\n",
"\n",
"```bash\n",
"docker run \\\n",
" --rm \\\n",
" --gpus all \\\n",
" --ipc=host \\\n",
" -p 8080:80 \\\n",
" -v ~/.cache/huggingface/hub:/data \\\n",
" -e HF_API_TOKEN=${HF_API_TOKEN} \\\n",
" ghcr.io/huggingface/text-generation-inference:0.9 \\\n",
" --hostname 0.0.0.0 \\\n",
" --model-id meta-llama/Llama-2-13b-chat-hf \\\n",
" --quantize bitsandbytes \\\n",
" --num-shard 4\n",
"```\n",
"\n",
"This works on a machine with 4 x RTX 3080ti cards, for example. Adjust the `--num_shard` value to the number of GPUs available. The `HF_API_TOKEN` environment variable holds the Hugging Face API token."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "238095fd",
"metadata": {},
"outputs": [],
"source": [
"# !pip3 install text-generation"
]
},
{
"cell_type": "markdown",
"id": "79c4ace9",
"metadata": {},
"source": [
"Create a `HuggingFaceTextGenInference` instance that connects to the local inference server and wrap it into `Llama2Chat`."
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "7a9f6de2",
"metadata": {},
"outputs": [],
"source": [
"from langchain.llms import HuggingFaceTextGenInference\n",
"\n",
"llm = HuggingFaceTextGenInference(\n",
" inference_server_url=\"http://127.0.0.1:8080/\",\n",
" max_new_tokens=512,\n",
" top_k=50,\n",
" temperature=0.1,\n",
" repetition_penalty=1.03,\n",
")\n",
"\n",
"model = Llama2Chat(llm=llm)"
]
},
{
"cell_type": "markdown",
"id": "4f646a2b",
"metadata": {},
"source": [
"Then you are ready to use the chat `model` together with `prompt_template` and conversation `memory` in an `LLMChain`."
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "54b5d1d1",
"metadata": {},
"outputs": [],
"source": [
"memory = ConversationBufferMemory(memory_key=\"chat_history\", return_messages=True)\n",
"chain = LLMChain(llm=model, prompt=prompt_template, memory=memory)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "e6717947",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Sure, I'd be happy to help! Here are a few popular locations to consider visiting in Vienna:\n",
"\n",
"1. Schönbrunn Palace\n",
"2. St. Stephen's Cathedral\n",
"3. Hofburg Palace\n",
"4. Belvedere Palace\n",
"5. Prater Park\n",
"6. Vienna State Opera\n",
"7. Albertina Museum\n",
"8. Museum of Natural History\n",
"9. Kunsthistorisches Museum\n",
"10. Ringstrasse\n"
]
}
],
"source": [
"print(\n",
" chain.run(\n",
" text=\"What can I see in Vienna? Propose a few locations. Names only, no details.\"\n",
" )\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "17bf10d5",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Certainly! St. Stephen's Cathedral (Stephansdom) is one of the most recognizable landmarks in Vienna and a must-see attraction for visitors. This stunning Gothic cathedral is located in the heart of the city and is known for its intricate stone carvings, colorful stained glass windows, and impressive dome.\n",
"\n",
"The cathedral was built in the 12th century and has been the site of many important events throughout history, including the coronation of Holy Roman emperors and the funeral of Mozart. Today, it is still an active place of worship and offers guided tours, concerts, and special events. Visitors can climb up the south tower for panoramic views of the city or attend a service to experience the beautiful music and chanting.\n"
]
}
],
"source": [
"print(chain.run(text=\"Tell me more about #2.\"))"
]
},
{
"cell_type": "markdown",
"id": "2a297e09",
"metadata": {},
"source": [
"## Chat with Llama-2 via `LlamaCPP` LLM"
]
},
{
"cell_type": "markdown",
"id": "52c1a0b9",
"metadata": {},
"source": [
"For using a Llama-2 chat model with a [LlamaCPP](https://python.langchain.com/docs/integrations/llms/llamacpp) `LMM`, install the `llama-cpp-python` library using [these installation instructions](https://python.langchain.com/docs/integrations/llms/llamacpp#installation). The following example uses a quantized [llama-2-7b-chat.Q4_0.gguf](https://huggingface.co/TheBloke/Llama-2-7b-Chat-GGUF/resolve/main/llama-2-7b-chat.Q4_0.gguf) model stored locally at `~/Models/llama-2-7b-chat.Q4_0.gguf`. \n",
"\n",
"After creating a `LlamaCpp` instance, the `llm` is again wrapped into `Llama2Chat`"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "07c0d04e",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"llama_model_loader: loaded meta data with 19 key-value pairs and 291 tensors from /home/martin/Models/llama-2-7b-chat.Q4_0.gguf (version GGUF V2)\n",
"llama_model_loader: - tensor 0: token_embd.weight q4_0 [ 4096, 32000, 1, 1 ]\n",
"llama_model_loader: - tensor 1: blk.0.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 2: blk.0.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 3: blk.0.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 4: blk.0.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 5: blk.0.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 6: blk.0.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 7: blk.0.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 8: blk.0.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 9: blk.0.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 10: blk.1.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 11: blk.1.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 12: blk.1.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 13: blk.1.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 14: blk.1.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 15: blk.1.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 16: blk.1.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 17: blk.1.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 18: blk.1.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 19: blk.10.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 20: blk.10.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 21: blk.10.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 22: blk.10.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 23: blk.10.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 24: blk.10.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 25: blk.10.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 26: blk.10.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 27: blk.10.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 28: blk.11.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 29: blk.11.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 30: blk.11.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 31: blk.11.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 32: blk.11.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 33: blk.11.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 34: blk.11.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 35: blk.11.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 36: blk.11.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 37: blk.12.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 38: blk.12.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 39: blk.12.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 40: blk.12.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 41: blk.12.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 42: blk.12.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 43: blk.12.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 44: blk.12.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 45: blk.12.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 46: blk.13.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 47: blk.13.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 48: blk.13.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 49: blk.13.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 50: blk.13.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 51: blk.13.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 52: blk.13.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 53: blk.13.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 54: blk.13.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 55: blk.14.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 56: blk.14.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 57: blk.14.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 58: blk.14.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 59: blk.14.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 60: blk.14.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 61: blk.14.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 62: blk.14.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 63: blk.14.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 64: blk.15.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 65: blk.15.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 66: blk.15.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 67: blk.15.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 68: blk.15.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 69: blk.15.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 70: blk.15.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 71: blk.15.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 72: blk.15.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 73: blk.16.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 74: blk.16.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 75: blk.16.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 76: blk.16.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 77: blk.16.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 78: blk.16.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 79: blk.16.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 80: blk.16.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 81: blk.16.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 82: blk.17.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 83: blk.17.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 84: blk.17.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 85: blk.17.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 86: blk.17.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 87: blk.17.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 88: blk.17.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 89: blk.17.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 90: blk.17.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 91: blk.18.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 92: blk.18.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 93: blk.18.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 94: blk.18.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 95: blk.18.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 96: blk.18.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 97: blk.18.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 98: blk.18.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 99: blk.18.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 100: blk.19.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 101: blk.19.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 102: blk.19.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 103: blk.19.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 104: blk.19.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 105: blk.19.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 106: blk.19.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 107: blk.19.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 108: blk.19.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 109: blk.2.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 110: blk.2.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 111: blk.2.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 112: blk.2.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 113: blk.2.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 114: blk.2.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 115: blk.2.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 116: blk.2.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 117: blk.2.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 118: blk.20.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 119: blk.20.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 120: blk.20.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 121: blk.20.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 122: blk.20.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 123: blk.20.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 124: blk.20.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 125: blk.20.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 126: blk.20.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 127: blk.21.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 128: blk.21.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 129: blk.21.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 130: blk.21.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 131: blk.21.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 132: blk.21.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 133: blk.21.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 134: blk.21.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 135: blk.21.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 136: blk.22.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 137: blk.22.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 138: blk.22.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 139: blk.22.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 140: blk.22.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 141: blk.22.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 142: blk.22.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 143: blk.22.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 144: blk.22.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 145: blk.23.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 146: blk.23.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 147: blk.23.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 148: blk.23.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 149: blk.23.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 150: blk.23.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 151: blk.23.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 152: blk.23.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 153: blk.23.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 154: blk.3.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 155: blk.3.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 156: blk.3.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 157: blk.3.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 158: blk.3.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 159: blk.3.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 160: blk.3.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 161: blk.3.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 162: blk.3.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 163: blk.4.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 164: blk.4.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 165: blk.4.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 166: blk.4.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 167: blk.4.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 168: blk.4.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 169: blk.4.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 170: blk.4.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 171: blk.4.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 172: blk.5.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 173: blk.5.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 174: blk.5.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 175: blk.5.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 176: blk.5.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 177: blk.5.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 178: blk.5.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 179: blk.5.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 180: blk.5.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 181: blk.6.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 182: blk.6.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 183: blk.6.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 184: blk.6.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 185: blk.6.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 186: blk.6.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 187: blk.6.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 188: blk.6.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 189: blk.6.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 190: blk.7.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 191: blk.7.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 192: blk.7.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 193: blk.7.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 194: blk.7.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 195: blk.7.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 196: blk.7.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 197: blk.7.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 198: blk.7.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 199: blk.8.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 200: blk.8.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 201: blk.8.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 202: blk.8.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 203: blk.8.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 204: blk.8.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 205: blk.8.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 206: blk.8.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 207: blk.8.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 208: blk.9.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 209: blk.9.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 210: blk.9.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 211: blk.9.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 212: blk.9.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 213: blk.9.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 214: blk.9.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 215: blk.9.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 216: blk.9.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 217: output.weight q6_K [ 4096, 32000, 1, 1 ]\n",
"llama_model_loader: - tensor 218: blk.24.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 219: blk.24.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 220: blk.24.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 221: blk.24.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 222: blk.24.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 223: blk.24.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 224: blk.24.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 225: blk.24.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 226: blk.24.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 227: blk.25.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 228: blk.25.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 229: blk.25.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 230: blk.25.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 231: blk.25.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 232: blk.25.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 233: blk.25.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 234: blk.25.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 235: blk.25.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 236: blk.26.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 237: blk.26.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 238: blk.26.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 239: blk.26.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 240: blk.26.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 241: blk.26.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 242: blk.26.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 243: blk.26.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 244: blk.26.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 245: blk.27.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 246: blk.27.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 247: blk.27.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 248: blk.27.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 249: blk.27.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 250: blk.27.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 251: blk.27.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 252: blk.27.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 253: blk.27.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 254: blk.28.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 255: blk.28.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 256: blk.28.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 257: blk.28.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 258: blk.28.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 259: blk.28.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 260: blk.28.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 261: blk.28.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 262: blk.28.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 263: blk.29.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 264: blk.29.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 265: blk.29.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 266: blk.29.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 267: blk.29.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 268: blk.29.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 269: blk.29.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 270: blk.29.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 271: blk.29.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 272: blk.30.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 273: blk.30.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 274: blk.30.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 275: blk.30.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 276: blk.30.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 277: blk.30.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 278: blk.30.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 279: blk.30.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 280: blk.30.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 281: blk.31.attn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 282: blk.31.ffn_down.weight q4_0 [ 11008, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 283: blk.31.ffn_gate.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 284: blk.31.ffn_up.weight q4_0 [ 4096, 11008, 1, 1 ]\n",
"llama_model_loader: - tensor 285: blk.31.ffn_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - tensor 286: blk.31.attn_k.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 287: blk.31.attn_output.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 288: blk.31.attn_q.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 289: blk.31.attn_v.weight q4_0 [ 4096, 4096, 1, 1 ]\n",
"llama_model_loader: - tensor 290: output_norm.weight f32 [ 4096, 1, 1, 1 ]\n",
"llama_model_loader: - kv 0: general.architecture str \n",
"llama_model_loader: - kv 1: general.name str \n",
"llama_model_loader: - kv 2: llama.context_length u32 \n",
"llama_model_loader: - kv 3: llama.embedding_length u32 \n",
"llama_model_loader: - kv 4: llama.block_count u32 \n",
"llama_model_loader: - kv 5: llama.feed_forward_length u32 \n",
"llama_model_loader: - kv 6: llama.rope.dimension_count u32 \n",
"llama_model_loader: - kv 7: llama.attention.head_count u32 \n",
"llama_model_loader: - kv 8: llama.attention.head_count_kv u32 \n",
"llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 \n",
"llama_model_loader: - kv 10: general.file_type u32 \n",
"llama_model_loader: - kv 11: tokenizer.ggml.model str \n",
"llama_model_loader: - kv 12: tokenizer.ggml.tokens arr \n",
"llama_model_loader: - kv 13: tokenizer.ggml.scores arr \n",
"llama_model_loader: - kv 14: tokenizer.ggml.token_type arr \n",
"llama_model_loader: - kv 15: tokenizer.ggml.bos_token_id u32 \n",
"llama_model_loader: - kv 16: tokenizer.ggml.eos_token_id u32 \n",
"llama_model_loader: - kv 17: tokenizer.ggml.unknown_token_id u32 \n",
"llama_model_loader: - kv 18: general.quantization_version u32 \n",
"llama_model_loader: - type f32: 65 tensors\n",
"llama_model_loader: - type q4_0: 225 tensors\n",
"llama_model_loader: - type q6_K: 1 tensors\n",
"llm_load_vocab: special tokens definition check successful ( 259/32000 ).\n",
"llm_load_print_meta: format = GGUF V2\n",
"llm_load_print_meta: arch = llama\n",
"llm_load_print_meta: vocab type = SPM\n",
"llm_load_print_meta: n_vocab = 32000\n",
"llm_load_print_meta: n_merges = 0\n",
"llm_load_print_meta: n_ctx_train = 4096\n",
"llm_load_print_meta: n_embd = 4096\n",
"llm_load_print_meta: n_head = 32\n",
"llm_load_print_meta: n_head_kv = 32\n",
"llm_load_print_meta: n_layer = 32\n",
"llm_load_print_meta: n_rot = 128\n",
"llm_load_print_meta: n_gqa = 1\n",
"llm_load_print_meta: f_norm_eps = 0.0e+00\n",
"llm_load_print_meta: f_norm_rms_eps = 1.0e-06\n",
"llm_load_print_meta: f_clamp_kqv = 0.0e+00\n",
"llm_load_print_meta: f_max_alibi_bias = 0.0e+00\n",
"llm_load_print_meta: n_ff = 11008\n",
"llm_load_print_meta: rope scaling = linear\n",
"llm_load_print_meta: freq_base_train = 10000.0\n",
"llm_load_print_meta: freq_scale_train = 1\n",
"llm_load_print_meta: n_yarn_orig_ctx = 4096\n",
"llm_load_print_meta: rope_finetuned = unknown\n",
"llm_load_print_meta: model type = 7B\n",
"llm_load_print_meta: model ftype = mostly Q4_0\n",
"llm_load_print_meta: model params = 6.74 B\n",
"llm_load_print_meta: model size = 3.56 GiB (4.54 BPW) \n",
"llm_load_print_meta: general.name = LLaMA v2\n",
"llm_load_print_meta: BOS token = 1 '<s>'\n",
"llm_load_print_meta: EOS token = 2 '</s>'\n",
"llm_load_print_meta: UNK token = 0 '<unk>'\n",
"llm_load_print_meta: LF token = 13 '<0x0A>'\n",
"llm_load_tensors: ggml ctx size = 0.11 MB\n",
"llm_load_tensors: mem required = 3647.97 MB\n",
"..................................................................................................\n",
"llama_new_context_with_model: n_ctx = 512\n",
"llama_new_context_with_model: freq_base = 10000.0\n",
"llama_new_context_with_model: freq_scale = 1\n",
"llama_new_context_with_model: kv self size = 256.00 MB\n",
"llama_build_graph: non-view tensors processed: 740/740\n",
"llama_new_context_with_model: compute buffer total size = 2.66 MB\n",
"AVX = 1 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 0 | AVX512_VNNI = 1 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | \n"
]
}
],
"source": [
"from os.path import expanduser\n",
"\n",
"from langchain.llms import LlamaCpp\n",
"\n",
"model_path = expanduser(\"~/Models/llama-2-7b-chat.Q4_0.gguf\")\n",
"\n",
"llm = LlamaCpp(\n",
" model_path=model_path,\n",
" streaming=False,\n",
")\n",
"model = Llama2Chat(llm=llm)"
]
},
{
"cell_type": "markdown",
"id": "50498d96",
"metadata": {},
"source": [
"and used in the same way as in the previous example."
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "90782b96",
"metadata": {},
"outputs": [],
"source": [
"memory = ConversationBufferMemory(memory_key=\"chat_history\", return_messages=True)\n",
"chain = LLMChain(llm=model, prompt=prompt_template, memory=memory)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "2160b26d",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" Of course! Vienna is a beautiful city with a rich history and culture. Here are some of the top tourist attractions you might want to consider visiting:\n",
"1. Schönbrunn Palace\n",
"2. St. Stephen's Cathedral\n",
"3. Hofburg Palace\n",
"4. Belvedere Palace\n",
"5. Prater Park\n",
"6. MuseumsQuartier\n",
"7. Ringstrasse\n",
"8. Vienna State Opera\n",
"9. Kunsthistorisches Museum\n",
"10. Imperial Palace\n",
"\n",
"These are just a few of the many amazing places to see in Vienna. Each one has its own unique history and charm, so I hope you enjoy exploring this beautiful city!\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n",
"llama_print_timings: load time = 250.46 ms\n",
"llama_print_timings: sample time = 56.40 ms / 144 runs ( 0.39 ms per token, 2553.37 tokens per second)\n",
"llama_print_timings: prompt eval time = 1444.25 ms / 47 tokens ( 30.73 ms per token, 32.54 tokens per second)\n",
"llama_print_timings: eval time = 8832.02 ms / 143 runs ( 61.76 ms per token, 16.19 tokens per second)\n",
"llama_print_timings: total time = 10645.94 ms\n"
]
}
],
"source": [
"print(\n",
" chain.run(\n",
" text=\"What can I see in Vienna? Propose a few locations. Names only, no details.\"\n",
" )\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "d9ce06e3",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Llama.generate: prefix-match hit\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
" Of course! St. Stephen's Cathedral (also known as Stephansdom) is a stunning Gothic-style cathedral located in the heart of Vienna, Austria. It is one of the most recognizable landmarks in the city and is considered a symbol of Vienna.\n",
"Here are some interesting facts about St. Stephen's Cathedral:\n",
"1. History: The construction of St. Stephen's Cathedral began in the 12th century on the site of a former Romanesque church, and it took over 600 years to complete. The cathedral has been renovated and expanded several times throughout its history, with the most significant renovation taking place in the 19th century.\n",
"2. Architecture: St. Stephen's Cathedral is built in the Gothic style, characterized by its tall spires, pointed arches, and intricate stone carvings. The cathedral features a mix of Romanesque, Gothic, and Baroque elements, making it a unique blend of styles.\n",
"3. Design: The cathedral's design is based on the plan of a cross with a long nave and two shorter arms extending from it. The main altar is\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n",
"llama_print_timings: load time = 250.46 ms\n",
"llama_print_timings: sample time = 100.60 ms / 256 runs ( 0.39 ms per token, 2544.73 tokens per second)\n",
"llama_print_timings: prompt eval time = 5128.71 ms / 160 tokens ( 32.05 ms per token, 31.20 tokens per second)\n",
"llama_print_timings: eval time = 16193.02 ms / 255 runs ( 63.50 ms per token, 15.75 tokens per second)\n",
"llama_print_timings: total time = 21988.57 ms\n"
]
}
],
"source": [
"print(chain.run(text=\"Tell me more about #2.\"))"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.18"
}
},
"nbformat": 4,
"nbformat_minor": 5
}

26
libs/experimental/langchain_experimental/chat_models/__init__.py
Unescape Escape View File

@ -0,0 +1,26 @@
"""**Chat Models** are a variation on language models.
While Chat Models use language models under the hood, the interface they expose
is a bit different. Rather than expose a "text in, text out" API, they expose
an interface where "chat messages" are the inputs and outputs.
**Class hierarchy:**
.. code-block::
BaseLanguageModel --> BaseChatModel --> <name> # Examples: ChatOpenAI, ChatGooglePalm
**Main helpers:**
.. code-block::
AIMessage, BaseMessage, HumanMessage
""" # noqa: E501
from langchain_experimental.chat_models.llm_wrapper import Llama2Chat, Orca, Vicuna
__all__ = [
"Llama2Chat",
"Orca",
"Vicuna",
]

163
libs/experimental/langchain_experimental/chat_models/llm_wrapper.py
Unescape Escape View File

@ -0,0 +1,163 @@
"""Generic Wrapper for chat LLMs, with sample implementations
for Llama-2-chat, Llama-2-instruct and Vicuna models.
"""
from typing import Any, List, Optional
from langchain.callbacks.manager import (
AsyncCallbackManagerForLLMRun,
CallbackManagerForLLMRun,
)
from langchain.chat_models.base import BaseChatModel
from langchain.llms.base import LLM
from langchain.schema import (
AIMessage,
BaseMessage,
ChatGeneration,
ChatResult,
HumanMessage,
LLMResult,
SystemMessage,
)
DEFAULT_SYSTEM_PROMPT = """You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.
If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.""" # noqa: E501
class ChatWrapper(BaseChatModel):
llm: LLM
sys_beg: str
sys_end: str
ai_n_beg: str
ai_n_end: str
usr_n_beg: str
usr_n_end: str
usr_0_beg: Optional[str] = None
usr_0_end: Optional[str] = None
system_message: SystemMessage = SystemMessage(content=DEFAULT_SYSTEM_PROMPT)
def _generate(
self,
messages: List[BaseMessage],
stop: Optional[List[str]] = None,
run_manager: Optional[CallbackManagerForLLMRun] = None,
**kwargs: Any,
) -> ChatResult:
llm_input = self._to_chat_prompt(messages)
llm_result = self.llm._generate(
prompts=[llm_input], stop=stop, run_manager=run_manager, **kwargs
)
return self._to_chat_result(llm_result)
async def _agenerate(
self,
messages: List[BaseMessage],
stop: Optional[List[str]] = None,
run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
**kwargs: Any,
) -> ChatResult:
llm_input = self._to_chat_prompt(messages)
llm_result = await self.llm._agenerate(
prompts=[llm_input], stop=stop, run_manager=run_manager, **kwargs
)
return self._to_chat_result(llm_result)
def _to_chat_prompt(
self,
messages: List[BaseMessage],
) -> str:
"""Convert a list of messages into a prompt format expected by wrapped LLM."""
if not messages:
raise ValueError("at least one HumanMessage must be provided")
if not isinstance(messages[0], SystemMessage):
messages = [self.system_message] + messages
if not isinstance(messages[1], HumanMessage):
raise ValueError(
"messages list must start with a SystemMessage or UserMessage"
)
if not isinstance(messages[-1], HumanMessage):
raise ValueError("last message must be a HumanMessage")
prompt_parts = []
if self.usr_0_beg is None:
self.usr_0_beg = self.usr_n_beg
if self.usr_0_end is None:
self.usr_0_end = self.usr_n_end
prompt_parts.append(self.sys_beg + messages[0].content + self.sys_end)
prompt_parts.append(self.usr_0_beg + messages[1].content + self.usr_0_end)
for ai_message, human_message in zip(messages[2::2], messages[3::2]):
if not isinstance(ai_message, AIMessage) or not isinstance(
human_message, HumanMessage
):
raise ValueError(
"messages must be alternating human- and ai-messages, "
"optionally prepended by a system message"
)
prompt_parts.append(self.ai_n_beg + ai_message.content + self.ai_n_end)
prompt_parts.append(self.usr_n_beg + human_message.content + self.usr_n_end)
return "".join(prompt_parts)
@staticmethod
def _to_chat_result(llm_result: LLMResult) -> ChatResult:
chat_generations = []
for g in llm_result.generations[0]:
chat_generation = ChatGeneration(
message=AIMessage(content=g.text), generation_info=g.generation_info
)
chat_generations.append(chat_generation)
return ChatResult(
generations=chat_generations, llm_output=llm_result.llm_output
)
class Llama2Chat(ChatWrapper):
@property
def _llm_type(self) -> str:
return "llama-2-chat"
sys_beg: str = "<s>[INST] <<SYS>>\n"
sys_end: str = "\n<</SYS>>\n\n"
ai_n_beg: str = " "
ai_n_end: str = " </s>"
usr_n_beg: str = "<s>[INST] "
usr_n_end: str = " [/INST]"
usr_0_beg: str = ""
usr_0_end: str = " [/INST]"
class Orca(ChatWrapper):
@property
def _llm_type(self) -> str:
return "orca-style"
sys_beg: str = "### System:\n"
sys_end: str = "\n\n"
ai_n_beg: str = "### Assistant:\n"
ai_n_end: str = "\n\n"
usr_n_beg: str = "### User:\n"
usr_n_end: str = "\n\n"
class Vicuna(ChatWrapper):
@property
def _llm_type(self) -> str:
return "vicuna-style"
sys_beg: str = ""
sys_end: str = " "
ai_n_beg: str = "ASSISTANT: "
ai_n_end: str = " </s>"
usr_n_beg: str = "USER: "
usr_n_end: str = " "

24
libs/experimental/poetry.lock generated
Unescape Escape View File

@ -2806,6 +2806,24 @@ tomli = {version = ">=1.0.0", markers = "python_version < \"3.11\""}
[package.extras]
testing = ["argcomplete", "attrs (>=19.2.0)", "hypothesis (>=3.56)", "mock", "nose", "pygments (>=2.7.2)", "requests", "setuptools", "xmlschema"]
[[package]]
name = "pytest-asyncio"
version = "0.20.3"
description = "Pytest support for asyncio"
optional = false
python-versions = ">=3.7"
files = [
{file = "pytest-asyncio-0.20.3.tar.gz", hash = "sha256:83cbf01169ce3e8eb71c6c278ccb0574d1a7a3bb8eaaf5e50e0ad342afb33b36"},
{file = "pytest_asyncio-0.20.3-py3-none-any.whl", hash = "sha256:f129998b209d04fcc65c96fc85c11e5316738358909a8399e93be553d7656442"},
]
[package.dependencies]
pytest = ">=6.1.0"
[package.extras]
docs = ["sphinx (>=5.3)", "sphinx-rtd-theme (>=1.0)"]
testing = ["coverage (>=6.2)", "flaky (>=3.5.0)", "hypothesis (>=5.7.1)", "mypy (>=0.931)", "pytest-trio (>=0.7.0)"]
[[package]]
name = "python-dateutil"
version = "2.8.2"
@ -3871,9 +3889,7 @@ python-versions = ">=3.7"
files = [
{file = "SQLAlchemy-2.0.23-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:638c2c0b6b4661a4fd264f6fb804eccd392745c5887f9317feb64bb7cb03b3ea"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e3b5036aa326dc2df50cba3c958e29b291a80f604b1afa4c8ce73e78e1c9f01d"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:787af80107fb691934a01889ca8f82a44adedbf5ef3d6ad7d0f0b9ac557e0c34"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c14eba45983d2f48f7546bb32b47937ee2cafae353646295f0e99f35b14286ab"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:0666031df46b9badba9bed00092a1ffa3aa063a5e68fa244acd9f08070e936d3"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:89a01238fcb9a8af118eaad3ffcc5dedaacbd429dc6fdc43fe430d3a941ff965"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-win32.whl", hash = "sha256:cabafc7837b6cec61c0e1e5c6d14ef250b675fa9c3060ed8a7e38653bd732ff8"},
{file = "SQLAlchemy-2.0.23-cp310-cp310-win_amd64.whl", hash = "sha256:87a3d6b53c39cd173990de2f5f4b83431d534a74f0e2f88bd16eabb5667e65c6"},
@ -3910,9 +3926,7 @@ files = [
{file = "SQLAlchemy-2.0.23-cp38-cp38-win_amd64.whl", hash = "sha256:964971b52daab357d2c0875825e36584d58f536e920f2968df8d581054eada4b"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:616fe7bcff0a05098f64b4478b78ec2dfa03225c23734d83d6c169eb41a93e55"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0e680527245895aba86afbd5bef6c316831c02aa988d1aad83c47ffe92655e74"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9585b646ffb048c0250acc7dad92536591ffe35dba624bb8fd9b471e25212a35"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4895a63e2c271ffc7a81ea424b94060f7b3b03b4ea0cd58ab5bb676ed02f4221"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:cc1d21576f958c42d9aec68eba5c1a7d715e5fc07825a629015fe8e3b0657fb0"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:967c0b71156f793e6662dd839da54f884631755275ed71f1539c95bbada9aaab"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-win32.whl", hash = "sha256:0a8c6aa506893e25a04233bc721c6b6cf844bafd7250535abb56cb6cc1368884"},
{file = "SQLAlchemy-2.0.23-cp39-cp39-win_amd64.whl", hash = "sha256:f3420d00d2cb42432c1d0e44540ae83185ccbbc67a6054dcc8ab5387add6620b"},
@ -4871,4 +4885,4 @@ extended-testing = ["faker", "presidio-analyzer", "presidio-anonymizer", "senten
[metadata]
lock-version = "2.0"
python-versions = ">=3.8.1,<4.0"
content-hash = "b834d2b8bcfb0c10549937841a9c6838ca8fde99d23e6c6deb8a6e3f4f4e43af"
content-hash = "ba9be2e62d1507b2f370b4388604d8e3e5afb3d495691f12d15d0128f162539d"

1
libs/experimental/pyproject.toml
Unescape Escape View File

@ -34,6 +34,7 @@ setuptools = "^67.6.1"
# dependencies used for running tests (e.g., pytest, freezegun, response).
# Any dependencies that do not meet that criteria will be removed.
pytest = "^7.3.0"
pytest-asyncio = "^0.20.3"
[tool.poetry.group.test_integration]

0
libs/experimental/tests/unit_tests/chat_models/__init__.py
Unescape Escape View File

157
libs/experimental/tests/unit_tests/chat_models/test_llm_wrapper_llama2chat.py
Unescape Escape View File

@ -0,0 +1,157 @@
from typing import Any, List, Optional
import pytest
from langchain.callbacks.manager import (
AsyncCallbackManagerForLLMRun,
CallbackManagerForLLMRun,
)
from langchain.llms.base import LLM
from langchain.schema import AIMessage, HumanMessage, SystemMessage
from langchain_experimental.chat_models import Llama2Chat
from langchain_experimental.chat_models.llm_wrapper import DEFAULT_SYSTEM_PROMPT
class FakeLLM(LLM):
def _call(
self,
prompt: str,
stop: Optional[List[str]] = None,
run_manager: Optional[CallbackManagerForLLMRun] = None,
**kwargs: Any,
) -> str:
return prompt
async def _acall(
self,
prompt: str,
stop: Optional[List[str]] = None,
run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
**kwargs: Any,
) -> str:
return prompt
@property
def _llm_type(self) -> str:
return "fake-llm"
@pytest.fixture
def model() -> Llama2Chat:
return Llama2Chat(llm=FakeLLM())
@pytest.fixture
def model_cfg_sys_msg() -> Llama2Chat:
return Llama2Chat(llm=FakeLLM(), system_message=SystemMessage(content="sys-msg"))
def test_default_system_message(model: Llama2Chat) -> None:
messages = [HumanMessage(content="usr-msg-1")]
actual = model.predict_messages(messages).content # type: ignore
expected = (
f"<s>[INST] <<SYS>>\n{DEFAULT_SYSTEM_PROMPT}\n<</SYS>>\n\nusr-msg-1 [/INST]"
)
assert actual == expected
def test_configured_system_message(
model_cfg_sys_msg: Llama2Chat,
) -> None:
messages = [HumanMessage(content="usr-msg-1")]
actual = model_cfg_sys_msg.predict_messages(messages).content # type: ignore
expected = "<s>[INST] <<SYS>>\nsys-msg\n<</SYS>>\n\nusr-msg-1 [/INST]"
assert actual == expected
@pytest.mark.asyncio
async def test_configured_system_message_async(
model_cfg_sys_msg: Llama2Chat,
) -> None:
messages = [HumanMessage(content="usr-msg-1")]
actual = await model_cfg_sys_msg.apredict_messages(messages) # type: ignore
expected = "<s>[INST] <<SYS>>\nsys-msg\n<</SYS>>\n\nusr-msg-1 [/INST]"
assert actual.content == expected
def test_provided_system_message(
model_cfg_sys_msg: Llama2Chat,
) -> None:
messages = [
SystemMessage(content="custom-sys-msg"),
HumanMessage(content="usr-msg-1"),
]
actual = model_cfg_sys_msg.predict_messages(messages).content
expected = "<s>[INST] <<SYS>>\ncustom-sys-msg\n<</SYS>>\n\nusr-msg-1 [/INST]"
assert actual == expected
def test_human_ai_dialogue(model_cfg_sys_msg: Llama2Chat) -> None:
messages = [
HumanMessage(content="usr-msg-1"),
AIMessage(content="ai-msg-1"),
HumanMessage(content="usr-msg-2"),
AIMessage(content="ai-msg-2"),
HumanMessage(content="usr-msg-3"),
]
actual = model_cfg_sys_msg.predict_messages(messages).content
expected = (
"<s>[INST] <<SYS>>\nsys-msg\n<</SYS>>\n\nusr-msg-1 [/INST] ai-msg-1 </s>"
"<s>[INST] usr-msg-2 [/INST] ai-msg-2 </s><s>[INST] usr-msg-3 [/INST]"
)
assert actual == expected
def test_no_message(model: Llama2Chat) -> None:
with pytest.raises(ValueError) as info:
model.predict_messages([])
assert info.value.args[0] == "at least one HumanMessage must be provided"
def test_ai_message_first(model: Llama2Chat) -> None:
with pytest.raises(ValueError) as info:
model.predict_messages([AIMessage(content="ai-msg-1")])
assert (
info.value.args[0]
== "messages list must start with a SystemMessage or UserMessage"
)
def test_human_ai_messages_not_alternating(model: Llama2Chat) -> None:
messages = [
HumanMessage(content="usr-msg-1"),
HumanMessage(content="usr-msg-2"),
HumanMessage(content="ai-msg-1"),
]
with pytest.raises(ValueError) as info:
model.predict_messages(messages) # type: ignore
assert info.value.args[0] == (
"messages must be alternating human- and ai-messages, "
"optionally prepended by a system message"
)
def test_last_message_not_human_message(model: Llama2Chat) -> None:
messages = [
HumanMessage(content="usr-msg-1"),
AIMessage(content="ai-msg-1"),
]
with pytest.raises(ValueError) as info:
model.predict_messages(messages)
assert info.value.args[0] == "last message must be a HumanMessage"

29
libs/experimental/tests/unit_tests/chat_models/test_llm_wrapper_orca.py
Unescape Escape View File

@ -0,0 +1,29 @@
import pytest
from langchain.schema import AIMessage, HumanMessage, SystemMessage
from langchain_experimental.chat_models import Orca
from tests.unit_tests.chat_models.test_llm_wrapper_llama2chat import FakeLLM
@pytest.fixture
def model() -> Orca:
return Orca(llm=FakeLLM())
@pytest.fixture
def model_cfg_sys_msg() -> Orca:
return Orca(llm=FakeLLM(), system_message=SystemMessage(content="sys-msg"))
def test_prompt(model: Orca) -> None:
messages = [
SystemMessage(content="sys-msg"),
HumanMessage(content="usr-msg-1"),
AIMessage(content="ai-msg-1"),
HumanMessage(content="usr-msg-2"),
]
actual = model.predict_messages(messages).content # type: ignore
expected = "### System:\nsys-msg\n\n### User:\nusr-msg-1\n\n### Assistant:\nai-msg-1\n\n### User:\nusr-msg-2\n\n" # noqa: E501
assert actual == expected

29
libs/experimental/tests/unit_tests/chat_models/test_llm_wrapper_vicuna.py
Unescape Escape View File

@ -0,0 +1,29 @@
import pytest
from langchain.schema import AIMessage, HumanMessage, SystemMessage
from langchain_experimental.chat_models import Vicuna
from tests.unit_tests.chat_models.test_llm_wrapper_llama2chat import FakeLLM
@pytest.fixture
def model() -> Vicuna:
return Vicuna(llm=FakeLLM())
@pytest.fixture
def model_cfg_sys_msg() -> Vicuna:
return Vicuna(llm=FakeLLM(), system_message=SystemMessage(content="sys-msg"))
def test_prompt(model: Vicuna) -> None:
messages = [
SystemMessage(content="sys-msg"),
HumanMessage(content="usr-msg-1"),
AIMessage(content="ai-msg-1"),
HumanMessage(content="usr-msg-2"),
]
actual = model.predict_messages(messages).content # type: ignore
expected = "sys-msg USER: usr-msg-1 ASSISTANT: ai-msg-1 </s>USER: usr-msg-2 "
assert actual == expected
Write Preview
Loading…
Cancel
Save