Instructions to use CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421 with Transformers:

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

pipe = pipeline("text-generation", model="CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421")
model = AutoModelForCausalLM.from_pretrained("CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421

SGLang

How to use CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421 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 "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421 with Docker Model Runner:
```
docker model run hf.co/CMU-AIR2/code-ArithHardC12-curriculum-gh-mixture-240421
```

code-ArithHardC12-curriculum-gh-mixture-240421 / trainer_state.json

cterdam

Upload folder using huggingface_hub

cfab25c verified over 2 years ago

Raw

History Blame Contribute Delete

1.19 kB

	{
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 8.605851979345955,
	"eval_steps": 2500,
	"global_step": 5000,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 4.3,
	"grad_norm": 1.1691417694091797,
	"learning_rate": 1.1111111111111113e-05,
	"loss": 0.2129,
	"step": 2500
	},
	{
	"epoch": 4.3,
	"eval_loss": 0.333095908164978,
	"eval_runtime": 83.9342,
	"eval_samples_per_second": 11.914,
	"eval_steps_per_second": 2.979,
	"step": 2500
	},
	{
	"epoch": 8.61,
	"grad_norm": 0.1415511667728424,
	"learning_rate": 0.0,
	"loss": 0.0149,
	"step": 5000
	},
	{
	"epoch": 8.61,
	"eval_loss": 0.4397329092025757,
	"eval_runtime": 83.9924,
	"eval_samples_per_second": 11.906,
	"eval_steps_per_second": 2.976,
	"step": 5000
	}
	],
	"logging_steps": 2500,
	"max_steps": 5000,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 9,
	"save_steps": 2500,
	"total_flos": 1.258569996863275e+18,
	"train_batch_size": 4,
	"trial_name": null,
	"trial_params": null
	}