Instructions to use subset-data/falcon-testing with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use subset-data/falcon-testing with Transformers:

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

pipe = pipeline("text-generation", model="subset-data/falcon-testing", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("subset-data/falcon-testing", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use subset-data/falcon-testing with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/subset-data/falcon-testing

SGLang

How to use subset-data/falcon-testing 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 "subset-data/falcon-testing" \
    --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": "subset-data/falcon-testing",
		"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 "subset-data/falcon-testing" \
        --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": "subset-data/falcon-testing",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use subset-data/falcon-testing with Docker Model Runner:
```
docker model run hf.co/subset-data/falcon-testing
```

falcon-testing / last-checkpoint /trainer_state.json

bwebs

Training in progress, step 17

1d42776 about 3 years ago

Raw

History Blame Contribute Delete

2.1 kB

	{
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 4.25,
	"global_step": 17,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.25,
	"learning_rate": 0.0002,
	"loss": 1.7413,
	"step": 1
	},
	{
	"epoch": 0.5,
	"learning_rate": 0.0002,
	"loss": 1.7187,
	"step": 2
	},
	{
	"epoch": 0.75,
	"learning_rate": 0.0002,
	"loss": 1.7006,
	"step": 3
	},
	{
	"epoch": 1.0,
	"learning_rate": 0.0002,
	"loss": 1.5806,
	"step": 4
	},
	{
	"epoch": 1.25,
	"learning_rate": 0.0002,
	"loss": 1.5316,
	"step": 5
	},
	{
	"epoch": 1.5,
	"learning_rate": 0.0002,
	"loss": 1.4856,
	"step": 6
	},
	{
	"epoch": 1.75,
	"learning_rate": 0.0002,
	"loss": 1.3389,
	"step": 7
	},
	{
	"epoch": 2.0,
	"learning_rate": 0.0002,
	"loss": 1.2343,
	"step": 8
	},
	{
	"epoch": 2.25,
	"learning_rate": 0.0002,
	"loss": 1.1174,
	"step": 9
	},
	{
	"epoch": 2.5,
	"learning_rate": 0.0002,
	"loss": 0.9802,
	"step": 10
	},
	{
	"epoch": 2.75,
	"learning_rate": 0.0002,
	"loss": 0.8965,
	"step": 11
	},
	{
	"epoch": 3.0,
	"learning_rate": 0.0002,
	"loss": 0.7951,
	"step": 12
	},
	{
	"epoch": 3.25,
	"learning_rate": 0.0002,
	"loss": 0.6422,
	"step": 13
	},
	{
	"epoch": 3.5,
	"learning_rate": 0.0002,
	"loss": 0.6023,
	"step": 14
	},
	{
	"epoch": 3.75,
	"learning_rate": 0.0002,
	"loss": 0.5078,
	"step": 15
	},
	{
	"epoch": 4.0,
	"learning_rate": 0.0002,
	"loss": 0.5033,
	"step": 16
	},
	{
	"epoch": 4.25,
	"learning_rate": 0.0002,
	"loss": 0.4317,
	"step": 17
	}
	],
	"max_steps": 500,
	"num_train_epochs": 125,
	"total_flos": 2835286102376448.0,
	"trial_name": null,
	"trial_params": null
	}