base_model: Qwen/Qwen2.5-14B-Instruct

adapter: qlora
load_in_4bit: true
load_in_8bit: false

bf16: true
torch_dtype: bfloat16

bnb_4bit_quant_type: nf4
bnb_4bit_compute_dtype: bfloat16
bnb_4bit_use_double_quant: true

output_dir: ./outputs/mymodel

datasets:
  - path: fares-boutriga/DamorkDataSet1
    type: chat_template
    data_files:
      - damork_dataset.axolotl.train.jsonl

micro_batch_size: 1
gradient_accumulation_steps: 8
sequence_len: 2048
num_epochs: 1
learning_rate: 2e-4

optimizer: adamw_bnb_8bit
lr_scheduler: cosine
weight_decay: 0.0

lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
  - q_proj
  - k_proj
  - v_proj
  - o_proj
  - gate_proj
  - up_proj
  - down_proj

val_set_size: 0.0
train_on_inputs: false

save_safetensors: true
save_only_model: false

outputs/mymodel

This model is a fine-tuned version of Qwen/Qwen2.5-14B-Instruct on the fares-boutriga/DamorkDataSet1 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: 0.0002
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 8
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: cosine
lr_scheduler_warmup_steps: 2
training_steps: 40

Training results

Framework versions

PEFT 0.17.1
Transformers 4.57.0
Pytorch 2.7.1+cu126
Datasets 4.0.0
Tokenizers 0.22.1

fares-boutriga/Damork-tx-1

outputs/mymodel

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

fares-boutriga/
Damork-tx-1