Instructions to use giux78/zagreus-test-202000-sft-8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use giux78/zagreus-test-202000-sft-8 with Transformers:

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

pipe = pipeline("text-generation", model="giux78/zagreus-test-202000-sft-8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("giux78/zagreus-test-202000-sft-8")
model = AutoModelForCausalLM.from_pretrained("giux78/zagreus-test-202000-sft-8")
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

vLLM

How to use giux78/zagreus-test-202000-sft-8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "giux78/zagreus-test-202000-sft-8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "giux78/zagreus-test-202000-sft-8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/giux78/zagreus-test-202000-sft-8

SGLang

How to use giux78/zagreus-test-202000-sft-8 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 "giux78/zagreus-test-202000-sft-8" \
    --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": "giux78/zagreus-test-202000-sft-8",
		"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 "giux78/zagreus-test-202000-sft-8" \
        --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": "giux78/zagreus-test-202000-sft-8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use giux78/zagreus-test-202000-sft-8 with Docker Model Runner:
```
docker model run hf.co/giux78/zagreus-test-202000-sft-8
```

See axolotl config

axolotl version: 0.9.2

# ============= SFT PRODUCTION (4M ShareGPT) =============
base_model: giux78/zagreus-test-202000
strict: false
output_dir: ./ale_outputs/opendata-sft-prod
seed: 42

# ---- Dataset ----
datasets:
  - path: /leonardo_work/EUHPC_A04_045/training/opendata-1000000
    type: chat_template
    field_messages: conversation
    roles_to_train: ["assistant"]   # loss solo sui turni assistant
    train_on_eos: turn              # predici <|eot_id|> a fine risposta assistant

# (opzionale ma consigliato: cache pretokenizzata tra run)
dataset_prepared_path: ./ale_outputs/dataset_cache/opendata-sft

default_system_message: "Sei un assistente utile."

# ---- Chat template (Llama-3.2 style) ----
chat_template: llama3 

# ---- Training ----
sequence_len: 4096
sample_packing: true            # ON per efficienza
eval_sample_packing: true
pad_to_sequence_len: true

optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 1.5e-5
warmup_ratio: 0.03              # ~3% dei passi totali
weight_decay: 0.01
max_grad_norm: 1.0

# 32 GPU totali -> eff. batch = 1 * 8 * 32 = 256
micro_batch_size: 1
gradient_accumulation_steps: 8

num_epochs: 1.0                 # 1 epoca completa su 4M conv
# (alternativa: usa max_steps se vuoi fermarti prima)

# ---- Precisione & memoria ----
bf16: auto
flash_attention: true
gradient_checkpointing: true

# ---- Log/Eval/Save ----
logging_steps: 20
eval_strategy: steps
eval_steps: 150                 # ~7-8 eval/epoca
save_strategy: steps
save_steps: 300                 # ~3 checkpoint/epoca
save_total_limit: 4
val_set_size: 10000  
# (opzionale) val_set_size: 10000   # se vuoi split automatico dal dataset

# ---- FSDP multi-nodo ----
fsdp_config:
  fsdp_sharding_strategy: FULL_SHARD
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
  fsdp_backward_prefetch_policy: BACKWARD_PRE
  fsdp_state_dict_type: FULL_STATE_DICT

# ---- Token speciali (coerenti col tokenizer del base_model) ----
special_tokens:
  pad_token: <|end_of_text|>
  eos_token: <|end_of_text|>

ale_outputs/opendata-sft-prod

This model is a fine-tuned version of giux78/zagreus-test-202000 on the None dataset. It achieves the following results on the evaluation set:

Loss: 3.7225

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: 1.5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 32
gradient_accumulation_steps: 8
total_train_batch_size: 256
total_eval_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: 14
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0020	1	3.7876
3.6507	0.3017	150	3.7436
3.7259	0.6033	300	3.7246
3.6753	0.9050	450	3.7225

Framework versions

Transformers 4.56.2
Pytorch 2.5.1+cu121
Datasets 3.5.1
Tokenizers 0.22.1

Downloads last month: 4

Safetensors

Model size

0.4B params

Tensor type

BF16

Model tree for giux78/zagreus-test-202000-sft-8

Base model

giux78/zagreus-test-202000

Finetuned

(12)

this model