See axolotl config

axolotl version: 0.10.0

base_model: meta-llama/Llama-3.1-8B-Instruct
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false

datasets:
  - path: cfierro/simpleqa_wiki_ar_Llama-3.1-8B-Instruct
    type: completion
    field: text  # also available: field_prompt, field_completion
    split: test
dataset_prepared_path: /scratch/project/eu-25-39/knowledge-ft/axolotl/datasets/Llama-2-7b-chat/simpleqa-ar
val_set_size: 0.0
output_dir: /scratch/project/eu-25-39/knowledge-ft/axolotl/models/llama-3.1-8b-fft-simpleqa-ar
hub_model_id: llama-3.1-8b-fft-simpleqa-ar

sequence_len: 4096
# Since most examples are ~600 in length, then each "example" is actually 6-8 examples
# which means that batch_size in reality is batch_size * 6
sample_packing: true
eval_sample_packing: false

# No LoRA — full fine-tuning

wandb_project: knowledge-ft
wandb_entity: cfierro
wandb_watch:
wandb_name: llama-3.1-8b-fft-simpleqa-ar
wandb_log_model: "false"

# Multi-GPU settings
# micro_batch=1 to fit full FT in memory (ZeRO-3 on 4x A100 40GB)
# grad_accum=2 to keep same effective batch size (1 * 2 * 4 GPUs = 8)
gradient_accumulation_steps: 2
micro_batch_size: 1
#num_epochs: 3
max_steps: 3000

optimizer: adamw_bnb_8bit
lr_scheduler: constant
learning_rate: 1e-5

bf16: auto
tf32: false

gradient_checkpointing: true
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.03  # Same from SEAL
#eval_steps: 1000
save_steps: 1000
save_total_limit: 1
load_best_model_at_end: true
weight_decay: 0.0
special_tokens:
   pad_token: <|end_of_text|>

# DeepSpeed ZeRO Stage 3 - shards model weights, gradients, and optimizer across GPUs
deepspeed: deepspeed_configs/zero3.json

llama-3.1-8b-fft-simpleqa-ar

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the cfierro/simpleqa_wiki_ar_Llama-3.1-8B-Instruct dataset.

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• lr_scheduler_warmup_steps: 9
• training_steps: 3000

Training results

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu128
• Datasets 3.5.0
• Tokenizers 0.22.2