Instructions to use finalform/foamQwen2.5-7B-Instruct-Assistant with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use finalform/foamQwen2.5-7B-Instruct-Assistant with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-7B-Instruct")
model = PeftModel.from_pretrained(base_model, "finalform/foamQwen2.5-7B-Instruct-Assistant")

Transformers

How to use finalform/foamQwen2.5-7B-Instruct-Assistant with Transformers:

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

pipe = pipeline("text-generation", model="finalform/foamQwen2.5-7B-Instruct-Assistant")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("finalform/foamQwen2.5-7B-Instruct-Assistant", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use finalform/foamQwen2.5-7B-Instruct-Assistant with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "finalform/foamQwen2.5-7B-Instruct-Assistant"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finalform/foamQwen2.5-7B-Instruct-Assistant",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/finalform/foamQwen2.5-7B-Instruct-Assistant

SGLang

How to use finalform/foamQwen2.5-7B-Instruct-Assistant 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 "finalform/foamQwen2.5-7B-Instruct-Assistant" \
    --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": "finalform/foamQwen2.5-7B-Instruct-Assistant",
		"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 "finalform/foamQwen2.5-7B-Instruct-Assistant" \
        --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": "finalform/foamQwen2.5-7B-Instruct-Assistant",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use finalform/foamQwen2.5-7B-Instruct-Assistant with Docker Model Runner:
```
docker model run hf.co/finalform/foamQwen2.5-7B-Instruct-Assistant
```

foamQwen2.5-7B-Instruct-Assistant / checkpoint-52 /trainer_state.json

finalform

Upload folder using huggingface_hub

90431f1 verified 7 months ago

raw

history blame contribute delete

2.27 kB

Invalid JSON:Unexpected token 'N', ..."al_loss": NaN, "... is not valid JSON

	{
	"best_global_step": null,
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 1.0,
	"eval_steps": 500,
	"global_step": 52,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.1927710843373494,
	"grad_norm": 95.26730346679688,
	"learning_rate": 0.0004180909090909091,
	"loss": 22.4465,
	"mean_token_accuracy": 0.08969678990542888,
	"num_tokens": 1310720.0,
	"step": 10
	},
	{
	"epoch": 0.3855421686746988,
	"grad_norm": 0.0,
	"learning_rate": 0.0005103526968014265,
	"loss": 0.8012,
	"mean_token_accuracy": 0.13804710581898688,
	"num_tokens": 2621440.0,
	"step": 20
	},
	{
	"epoch": 0.5783132530120482,
	"grad_norm": 0.0,
	"learning_rate": 0.0005077286477833616,
	"loss": 452.9577,
	"mean_token_accuracy": 0.05160275483503938,
	"num_tokens": 3932160.0,
	"step": 30
	},
	{
	"epoch": 0.7710843373493976,
	"grad_norm": 0.0,
	"learning_rate": 0.0005031081504278389,
	"loss": 470.5136,
	"mean_token_accuracy": 0.03822226445190609,
	"num_tokens": 5242880.0,
	"step": 40
	},
	{
	"epoch": 0.963855421686747,
	"grad_norm": 26.303752899169922,
	"learning_rate": 0.0004965277770447238,
	"loss": 167.1384,
	"mean_token_accuracy": 0.057517293840646744,
	"num_tokens": 6553600.0,
	"step": 50
	},
	{
	"epoch": 1.0,
	"eval_loss": NaN,
	"eval_mean_token_accuracy": 0.13677339731378757,
	"eval_num_tokens": 6713344.0,
	"eval_runtime": 8.9806,
	"eval_samples_per_second": 41.089,
	"eval_steps_per_second": 5.233,
	"step": 52
	}
	],
	"logging_steps": 10,
	"max_steps": 364,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 7,
	"save_steps": 500,
	"stateful_callbacks": {
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": false
	},
	"attributes": {}
	}
	},
	"total_flos": 2.950418957829407e+17,
	"train_batch_size": 2,
	"trial_name": null,
	"trial_params": null
	}