Instructions to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
model = PeftModel.from_pretrained(base_model, "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k")

Transformers

How to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k with Transformers:

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

pipe = pipeline("text-generation", model="ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k

SGLang

How to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k 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 "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k" \
    --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": "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k",
		"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 "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k" \
        --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": "ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k with Docker Model Runner:
```
docker model run hf.co/ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k
```

crfTask-Llama-3.1-8B-Instruct-50k

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5774
F1 Micro: 0.9402
F1 Macro: 0.4933
F1 Weighted: 0.9471

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: 0.0003
train_batch_size: 16
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-07 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Micro	F1 Macro	F1 Weighted
7.2391	0.0776	60	1.4732	0.9236	0.4915	0.9288
6.1125	0.1551	120	1.1646	0.9502	0.5644	0.9527
5.1969	0.2327	180	1.0281	0.9369	0.5613	0.9423
4.7516	0.3103	240	0.9565	0.9369	0.6218	0.9430
4.4234	0.3878	300	0.8937	0.9568	0.6956	0.9594
4.0875	0.4654	360	0.8211	0.9502	0.6371	0.9525
3.6719	0.5430	420	0.7511	0.9402	0.5671	0.9461
3.2602	0.6206	480	0.6904	0.9468	0.5349	0.9526
3.0242	0.6981	540	0.6438	0.9369	0.5245	0.9441
2.6406	0.7757	600	0.6117	0.9402	0.4956	0.9465
2.5996	0.8533	660	0.5892	0.9369	0.4928	0.9438
2.6277	0.9308	720	0.5774	0.9402	0.4933	0.9471

Framework versions

PEFT 0.18.1
Transformers 4.51.0
Pytorch 2.8.0+cu128
Datasets 3.6.0
Tokenizers 0.21.0

Downloads last month: 2

Model tree for ferrazzipietro/crfTask-Llama-3.1-8B-Instruct-50k

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Adapter

(2418)

this model