Abstract

The Llama-3.1-8B-Instruct-w16a16-1node model is a Turkish legal instruction-tuned variant of Meta’s Llama-3.1-8B-Instruct, trained fully in bfloat16 precision using the default Tensorwise quantization scaling recipe. This model serves as the baseline reference for the “FSDP2 with Float8 Precision for Faster Training” study, where BF16 FSDP2 performance was compared against multiple FP8 mixed-precision configurations. The model was fine-tuned on the newmindai/EuroHPC-Legal dataset (Turkish legal multi-domain Q/A) using pure BF16 for both forward and backward computation, providing a stable, high-precision foundation against which FP8 speedups, efficiency gains, and convergence behavior were benchmarked. Trained on a single EuroHPC/BSC-class node with four NVIDIA H100 GPUs, this configuration achieved consistent convergence and served as the control setup for evaluating all Float8 recipe variants.

Experiment Context

This model was trained as part of our study for comparing FSDP2 with bfloat16 precision against FSDP2 with FP8 mixed precision bfp16-fp8. We used meta-llama/Llama-3.1-8B-Instruct. The model has been loaded using torch_dtype = bfloat16 and wrapped at once, also during forward/backward passes bfloat16 has been used for computations.

from torch.distributed._composable.fsdp import fully_shard

mesh_device_type = "cuda" if use_cuda else "cpu"
mesh = DeviceMesh(mesh_device_type, list(range(world_size)))
fsdp_kwargs = {
    "mesh": mesh,
    "reshard_after_forward": True,
}
model = fully_shard(model, **fsdp_kwargs)

Base Model Technical Specifications

Parameters: 8 Billion
Architecture Family: Llama 3.1
Maximum Position Embeddings: 131,072
Attention Heads: 32 (num_attention_heads)
Key-Value Heads: 8 (num_key_value_heads)
Hidden Layers: 32 (num_hidden_layers)
Hidden Size: 4,096 (hidden_size)
Intermediate Size: 14,336
Vocabulary Size: 128,256
Precision: bfloat16
RoPE Scaling: type llama3, factor = 8.0
RMS Norm Epsilon: 1e-05
Activation: SiLU

Training Methodology

Training Configuration

Model: meta-llama/Llama-3.1-8B-Instruct
Sequence Length: 4,096 (seq_len)
Epochs: 1
Per-Device Micro Batch Size: 2
Gradient Accumulation: 4
GPUs: 4 (via CUDA_VISIBLE_DEVICES=0,1,2,3)
dtype: bf16 && fp8=false
- Weights: bfloat16
- Activations: bfloat16
Optimizer: AdamW
- Learning Rate: 2e-5
- Weight Decay: 0.01
- Betas: (0.9, 0.95)
- Epsilon: 1e-8
LR Scheduler: Cosine; warmup = 10% (warmup_ratio=0.1) | also warmup_steps=100
Max Grad Norm: 1.0
Gradient Checkpointing: Enabled
Evaluation: every 5 steps (eval_steps=5, eval_samples=1000)
Checkpointing: every 10 steps; keep last 5; select best by eval_loss
Logging: every step to file; Weights & Biases in offline mode
Seed: 100
Distributed Training: torch.distributed.run (single node, multi-GPU)
- FSDP2 (Optimized Fully Sharded Data Parallel)

Setups

Precision: Used Half-precision bfloat16 as data type and for computation.
Hardware: HPC (EuroHPC/BSC-class) node with 4 × NVIDIA H100 GPUs.
Framework: PyTorch with torchrun for distributed training.

Dependencies

package	Version
Transformers	4.57.1
torch	2.9.0+cu128
accelerate	0.14.1
datasets	4.3.0
huggingface-hub	0.36.0
tensorboard	2.20.0
tensorboard-data-server	0.7.2
wandb	0.22.1

Job Details

model	Job ID	Runtime (mins)	Nodes	GPUs	Node-hour	GPU-hour	micro-batch	batch-size	gradient_accumulation	total_batch_size
Llama-3.1-8B-Instruct_w16a8_rw	31768103	115.75	1	4	1.929	7.716	2	2	4	32
Llama-3.1-8B-Instruct_w16a8_rw_with_gw_hp	31837629	109.00	1	4	1.816	7.266	2	2	4	32
Llama-3.1-8B-Instruct-w16a8-mxtw	31768031	64.00	1	4	1.066	4.266	2	2	4	32
Llama-3.1-8B-Instruct-w16a16-tw	31768074	138.75	1	4	2,312	9,25	2	2	4	32
Llama-3.1-8B-Instruct-w16a8-1node-bs8	31768093	123.75	1	4	2.062	8,250	2	2	4	32
Llama-3.1-8B-Instruct-w16a16-4nodes-bs32	31478433	31.75	4	4	2.117	8.467	4	4	8	512
Llama-3.1-8B-Instruct-w16a8-4nodes-bs32	31478468	39.75	4	4	2.650	10.600	4	4	8	512
Llama-3.1-8B-Instruct-w16a16-8nodes-bs32	31476914	22.00	8	4	2.933	11.733	4	4	8	1024
Llama-3.1-8B-Instruct-w16a8-8nodes-bs32	31476844	23.50	8	4	3.133	12.533	4	4	8	1024
Llama-3.1-8B-Instruct-w16a16-8nodes-bs64	31476914	22.00	8	4	2.933	11.733	4	8	8	1024
Llama-3.1-8B-Instruct-w16a8-8nodes-bs64	31476844	23.50	8	4	3.133	12.533	4	8	8	1024

Training Time Analysision

Model	Training Time (mins)	Memory Allocated (avg %)	GPU Utilization (avg %)	Speed vs bf16
Llama-3.1-8B-Instruct_w16a16-tw	138.75267	74.4189	56.6059%	_
Llama-3.1-8B-Instruct-w16a8-1node-bs8	123.75267	68.8982	97.5364%	12.11%
Llama-3.1-8B-Instruct_w16a8_rw	115.75364	69.6132	97.7689%	19.87%
Llama-3.1-8B-Instruct_w16a8_rw_with_gw_hp	109.00364	69.4806	97.3312%	27.33%
Llama-3.1-8B-Instruct-w16a8-mxtw	64.00328	68.8982	95.5661%	116.82%

All 15-models trained on(1Node,4Noes,8Nodes with both bfp16-fp8 && bfp16 configurations and fp8 recipes)

perplexity metric results for bfp16 && bfp16-fp8 configurations	Accuracy metric results for bfp16 && bfp16-fp8 configurations	Loss metric results for bfp16 && bfp16-fp8 configurations	Memory allocation for bfp16 && bfp16-fp8 configurations	Utilization for bfp16 && bfp16-fp8 configurations

Model	Max Loss (train)	Min Loss (train)	Avg Loss (train)	Final Loss (train)	± Std (train)	Max Loss (val)	Min Loss (val)	Avg Loss (val)	Final Loss (val)	± Std (val)
Llama-3.1-8B-Instruct-w16a8-rw	8	3.1682	0.5740	0.8118	0.6431	0.2746	1.0613	0.8394	0.8937	0.8394
Llama-3.1-8B-Instruct_w16a8_rw_with_gw_hp	8	3.1837	0.5763	0.8116	0.6420	0.2751	1.0599	0.8391	0.8933	0.8391
Llama-3.1-8B-Instruct-w16a8-mxtw	8	3.1983	0.5747	0.8115	0.6446	0.2758	1.0562	0.8384	0.8923	0.8384
Llama-3.1-8B-Instruct-w16a16-tw	8	3.1235	0.7203	0.9750	0.3344	0.7612	1.9113	0.8907	0.9831	0.1897
Llama-3.1-8B-Instruct-w16a8-1node-bs8	8	3.1661	0.7261	0.9804	0.3374	0.7672	1.9230	0.8948	0.9867	0.1906
Llama-3.1-8B-Instruct-w16a16-4nodes-bs32	32	3.2452	0.7414	0.9665	0.4844	0.7504	1.0538	0.8382	0.8844	0.0725
Llama-3.1-8B-Instruct-w16a8-4nodes-bs32	32	3.2840	0.7478	0.9748	0.4905	0.7581	1.0701	0.8430	0.8922	0.0764
Llama-3.1-8B-Instruct-w16a16-8nodes-bs32	32	3.2311	0.8448	1.1856	0.6434	0.8448	1.0257	0.8977	0.9460	0.0568
Llama-3.1-8B-Instruct-w16a8-8nodes-bs32	32	3.3003	0.8473	1.1866	0.6481	0.8473	1.0203	0.8992	0.9445	0.0539
Llama-3.1-8B-Instruct-w16a16-8nodes-bs64	64	3.2311	0.8448	1.1856	0.6434	0.8448	1.0257	0.8977	0.9460	0.0568
Llama-3.1-8B-Instruct-w16a8-8nodes-bs64	64	3.3003	0.8473	1.1866	0.6481	0.8473	1.0203	0.8992	0.9445	0.0539

Implementation

Gpu && Memory usage Profiling

To visualize the usage of the memory and gpu's vram/utilis and more you can profile them with pytorch-profiler tool or nsight systems profiler tool .

follow the steps to visualize with Pytorch-profiler :
1. pip install the versions that mentioned in the dependencies section of these libs tensorboard and tensorboard-data-server.
2. Visualize pytorch profiles by runing the command provided below.
```
tensorboard --logdir="./Llama-3.1-8B-Instruct_w16a16" --port="6006" 
```

follow the steps to visualize with nsight systems:
1. download the files on to you local machine.
2. download the full_version of the nsys profiler tool from here : https://developer.nvidia.com/nsight-systems/get-started.
3. open the nsys-rep profiles using the nsight systems gui .
also you can get a summary of the report using these command lines up to your porpuse :
```
nsys stats --report cuda_gpu_kern_sum /path/to/lama3.1_fp16_baseline.nsys-rep
```
```
nsys stats --report cuda_gpu_trace /path/to/lama3.1_fp16_baseline.nsys-rep
```
```
  nsys stats --report cuda_gpu_mem_size_sum /path/to/lama3.1_fp16_baseline.nsys-rep
```

Usage

Note: the final model has been saved in bfloat16 format. For inference, load the model in bfloat16 or float16 as shown below:

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "newmindai/Llama-3.1-8B-Instruct-w16a16-1node"
dtype = torch.bfloat16

tok = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=dtype,
    device_map="auto"
)

prompt = "Soru: Kişisel Verilerin Korunması Kanunu uyarınca hangi durumlarda açık rıza aranmaz? Cevap:"

inputs = tok(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
    out = model.generate(
        **inputs,
        max_new_tokens=256,
        do_sample=False
    )

print(tok.decode(out[0], skip_special_tokens=True))

Ethical Considerations and Disclaimers

Research & development purposes only; not a substitute for professional legal counsel.
Users must ensure compliance with data protection and sector regulations.
Potential biases may exist in domain data and model outputs.

Model & Data Card Metadata

Total Parameters: 8,030,261,248
Serialized Size (approx.): 16,060,522,496 bytes
Config precision: bfloat16
RoPE: llama3 scaling, factor 8.0

References and Citations

Base Model

@misc{meta_llama31_8b_instruct,
  title={Llama 3.1 8B Instruct},
  author={Meta AI},
  year={2024},
  howpublished={\url{https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct}}
}

Training Dataset

@misc{euro_hpc_legal,
  title={EuroHPC-Legal},
  author={newmindai},
  year={2025},
  howpublished={\url{https://huggingface.co/datasets/newmindai/EuroHPC-Legal}}
}