Instructions to use interview-eval/zephyr-7b-math-case-6-ep5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use interview-eval/zephyr-7b-math-case-6-ep5 with Transformers:

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

pipe = pipeline("text-generation", model="interview-eval/zephyr-7b-math-case-6-ep5")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("interview-eval/zephyr-7b-math-case-6-ep5")
model = AutoModelForCausalLM.from_pretrained("interview-eval/zephyr-7b-math-case-6-ep5", device_map="auto")
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 Settings

vLLM

How to use interview-eval/zephyr-7b-math-case-6-ep5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "interview-eval/zephyr-7b-math-case-6-ep5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "interview-eval/zephyr-7b-math-case-6-ep5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/interview-eval/zephyr-7b-math-case-6-ep5

SGLang

How to use interview-eval/zephyr-7b-math-case-6-ep5 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 "interview-eval/zephyr-7b-math-case-6-ep5" \
    --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": "interview-eval/zephyr-7b-math-case-6-ep5",
		"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 "interview-eval/zephyr-7b-math-case-6-ep5" \
        --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": "interview-eval/zephyr-7b-math-case-6-ep5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use interview-eval/zephyr-7b-math-case-6-ep5 with Docker Model Runner:
```
docker model run hf.co/interview-eval/zephyr-7b-math-case-6-ep5
```

zephyr-7b-math-case-6-ep5 / trainer_state.json

Juyoung Suk

Model save

a48772b verified almost 2 years ago

Raw

History Blame Contribute Delete

2.87 kB

	{
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 5.0,
	"eval_steps": 500,
	"global_step": 25,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.2,
	"grad_norm": 24.8895814604259,
	"learning_rate": 1e-05,
	"loss": 0.9347,
	"step": 1
	},
	{
	"epoch": 1.0,
	"grad_norm": 10.507921581694449,
	"learning_rate": 9.330127018922195e-06,
	"loss": 1.0402,
	"step": 5
	},
	{
	"epoch": 1.0,
	"eval_loss": 0.750956118106842,
	"eval_runtime": 13.4777,
	"eval_samples_per_second": 19.514,
	"eval_steps_per_second": 0.371,
	"step": 5
	},
	{
	"epoch": 2.0,
	"grad_norm": 3.2842066334321927,
	"learning_rate": 6.913417161825449e-06,
	"loss": 0.691,
	"step": 10
	},
	{
	"epoch": 2.0,
	"eval_loss": 0.5700622200965881,
	"eval_runtime": 13.3794,
	"eval_samples_per_second": 19.657,
	"eval_steps_per_second": 0.374,
	"step": 10
	},
	{
	"epoch": 3.0,
	"grad_norm": 3.0707933492232007,
	"learning_rate": 3.705904774487396e-06,
	"loss": 0.5239,
	"step": 15
	},
	{
	"epoch": 3.0,
	"eval_loss": 0.426382839679718,
	"eval_runtime": 13.3851,
	"eval_samples_per_second": 19.649,
	"eval_steps_per_second": 0.374,
	"step": 15
	},
	{
	"epoch": 4.0,
	"grad_norm": 3.1454642852295103,
	"learning_rate": 1.0332332985438248e-06,
	"loss": 0.3917,
	"step": 20
	},
	{
	"epoch": 4.0,
	"eval_loss": 0.33590540289878845,
	"eval_runtime": 13.3827,
	"eval_samples_per_second": 19.652,
	"eval_steps_per_second": 0.374,
	"step": 20
	},
	{
	"epoch": 5.0,
	"grad_norm": 2.732790702100083,
	"learning_rate": 0.0,
	"loss": 0.3221,
	"step": 25
	},
	{
	"epoch": 5.0,
	"eval_loss": 0.31167706847190857,
	"eval_runtime": 13.3175,
	"eval_samples_per_second": 19.749,
	"eval_steps_per_second": 0.375,
	"step": 25
	},
	{
	"epoch": 5.0,
	"step": 25,
	"total_flos": 10468982784000.0,
	"train_loss": 0.5895810270309448,
	"train_runtime": 388.4587,
	"train_samples_per_second": 3.385,
	"train_steps_per_second": 0.064
	}
	],
	"logging_steps": 5,
	"max_steps": 25,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 5,
	"save_steps": 100,
	"stateful_callbacks": {
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": true
	},
	"attributes": {}
	}
	},
	"total_flos": 10468982784000.0,
	"train_batch_size": 8,
	"trial_name": null,
	"trial_params": null
	}