Instructions to use Dynosaur/llama3-8b-math-sft-subtask-10-subset with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Dynosaur/llama3-8b-math-sft-subtask-10-subset with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Dynosaur/llama3-8b-math-sft-subtask-10-subset")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Dynosaur/llama3-8b-math-sft-subtask-10-subset")
model = AutoModelForCausalLM.from_pretrained("Dynosaur/llama3-8b-math-sft-subtask-10-subset")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Dynosaur/llama3-8b-math-sft-subtask-10-subset with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Dynosaur/llama3-8b-math-sft-subtask-10-subset"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Dynosaur/llama3-8b-math-sft-subtask-10-subset",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Dynosaur/llama3-8b-math-sft-subtask-10-subset

SGLang

How to use Dynosaur/llama3-8b-math-sft-subtask-10-subset with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Dynosaur/llama3-8b-math-sft-subtask-10-subset" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Dynosaur/llama3-8b-math-sft-subtask-10-subset",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Dynosaur/llama3-8b-math-sft-subtask-10-subset" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Dynosaur/llama3-8b-math-sft-subtask-10-subset",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Dynosaur/llama3-8b-math-sft-subtask-10-subset with Docker Model Runner:
```
docker model run hf.co/Dynosaur/llama3-8b-math-sft-subtask-10-subset
```

Upload 5 files

by hexuan21 - opened Sep 17, 2024

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+4750

-0

Files changed (5) hide show

README.md +61 -0
all_results.json +9 -0
generation_config.json +9 -0
train_results.json +9 -0
trainer_state.json +0 -0

README.md ADDED Viewed

	@@ -0,0 +1,61 @@

+---
+library_name: transformers
+license: llama3
+base_model: Dynosaur/llama3-8b-math-sft
+tags:
+- trl
+- sft
+- generated_from_trainer
+model-index:
+- name: llama3-8b-math-sft-subtask-10-subset
+  results: []
+---
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+# llama3-8b-math-sft-subtask-10-subset
+This model is a fine-tuned version of [Dynosaur/llama3-8b-math-sft](https://huggingface.co/Dynosaur/llama3-8b-math-sft) on an unknown dataset.
+## Model description
+More information needed
+## Intended uses & limitations
+More information needed
+## Training and evaluation data
+More information needed
+## Training procedure
+### Training hyperparameters
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 2
+### Training results
+### Framework versions
+- Transformers 4.44.2
+- Pytorch 2.4.1+cu121
+- Datasets 3.0.0
+- Tokenizers 0.19.1

all_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 1.9977298524404086,
+    "total_flos": 60745959505920.0,
+    "train_loss": 0.36325452142592635,
+    "train_runtime": 2345.5208,
+    "train_samples": 42274,
+    "train_samples_per_second": 36.047,
+    "train_steps_per_second": 0.281
+}

generation_config.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": 128001,
+  "max_length": 4096,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.44.2"
+}

train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 1.9977298524404086,
+    "total_flos": 60745959505920.0,
+    "train_loss": 0.36325452142592635,
+    "train_runtime": 2345.5208,
+    "train_samples": 42274,
+    "train_samples_per_second": 36.047,
+    "train_steps_per_second": 0.281
+}

trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff