mirror of
https://github.com/zenlm/engine.git
synced 2026-10-30 15:00:05 +00:00
c3be7b6be4
* Add model * Update loader * Fix llama config usage * Docs * Fix config no_rope_layers * Fix tie_word_embeddings default * Add chat template * Embed the chat templates * Fix embedding template * enable_thinking default true * Update examples * XML tools for smollm3 * Add smollm3 docs * Fix openai examples * Clippy --------- Co-authored-by: Eric Buehler <[email protected]>
44 lines
1.0 KiB
Python
44 lines
1.0 KiB
Python
from mistralrs import (
|
|
Runner,
|
|
Which,
|
|
ChatCompletionRequest,
|
|
Architecture,
|
|
AnyMoeConfig,
|
|
AnyMoeExpertType,
|
|
)
|
|
|
|
runner = Runner(
|
|
which=Which.Plain(
|
|
model_id="mistralai/Mistral-7B-Instruct-v0.1",
|
|
arch=Architecture.Mistral,
|
|
),
|
|
anymoe_config=AnyMoeConfig(
|
|
hidden_size=4096,
|
|
dataset_json="examples/amoe.json",
|
|
prefix="model.layers",
|
|
mlp="mlp",
|
|
expert_type=AnyMoeExpertType.FineTuned(),
|
|
lr=1e-3,
|
|
epochs=100,
|
|
batch_size=4,
|
|
model_ids=["HuggingFaceH4/zephyr-7b-beta"],
|
|
layers=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15],
|
|
loss_csv_path="loss.csv",
|
|
),
|
|
)
|
|
|
|
res = runner.send_chat_completion_request(
|
|
ChatCompletionRequest(
|
|
model="default",
|
|
messages=[
|
|
{"role": "user", "content": "Tell me a story about the Rust type system."}
|
|
],
|
|
max_tokens=256,
|
|
presence_penalty=1.0,
|
|
top_p=0.1,
|
|
temperature=0.1,
|
|
)
|
|
)
|
|
print(res.choices[0].message.content)
|
|
print(res.usage)
|