Instructions to use ErikDaska/lr_2e-04-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ErikDaska/lr_2e-04-sft with Transformers:

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

pipe = pipeline("text-generation", model="ErikDaska/lr_2e-04-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("ErikDaska/lr_2e-04-sft")
model = AutoModelForMultimodalLM.from_pretrained("ErikDaska/lr_2e-04-sft")
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 ErikDaska/lr_2e-04-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ErikDaska/lr_2e-04-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ErikDaska/lr_2e-04-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ErikDaska/lr_2e-04-sft

SGLang

How to use ErikDaska/lr_2e-04-sft 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 "ErikDaska/lr_2e-04-sft" \
    --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": "ErikDaska/lr_2e-04-sft",
		"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 "ErikDaska/lr_2e-04-sft" \
        --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": "ErikDaska/lr_2e-04-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ErikDaska/lr_2e-04-sft with Docker Model Runner:
```
docker model run hf.co/ErikDaska/lr_2e-04-sft
```

lr_2e-04-sft

This model is a fine-tuned version of ErikDaska/lr_2e-04_v2.0 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.7796

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: 5e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 0.03
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.0792	0.4444	200	0.9389
0.9923	0.8889	400	0.8776
0.9370	1.3333	600	0.8430
0.9298	1.7778	800	0.8208
0.8879	2.2222	1000	0.8084
0.8611	2.6667	1200	0.7948
0.8314	3.1111	1400	0.7884
0.7953	3.5556	1600	0.7852
0.8559	4.0	1800	0.7795
0.8336	4.4444	2000	0.7800
0.7885	4.8889	2200	0.7796
0.8192	5.0	2250	0.7796

Framework versions

Transformers 5.10.2
Pytorch 2.11.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

Downloads last month: 138

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for ErikDaska/lr_2e-04-sft

Base model

ErikDaska/lr_2e-04_v2.0

Finetuned

(1)

this model