adapter: lora
base_model: mistralai/Ministral-3-8B-Instruct-2512-BF16
bf16: true
datasets:
- path: AI-AgentSafa/dataset
  type: alpaca
gradient_accumulation_steps: 2
learning_rate: 0.0002
load_in_4bit: false
lora_alpha: 128
lora_dropout: 0.05
lora_r: 64
lora_target_modules:
- q_proj
- v_proj
- k_proj
- o_proj
- gate_proj
- down_proj
- up_proj
micro_batch_size: 4
num_epochs: 15
optimizer: paged_adamw_32bit
output_dir: /workspace/fine-tuning/outputs/ministral8b-tsql
sequence_len: 4096
train_on_inputs: false

workspace/fine-tuning/outputs/ministral8b-tsql

This model is a fine-tuned version of mistralai/Ministral-3-8B-Instruct-2512-BF16 on the AI-AgentSafa/dataset 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: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
optimizer: Use OptimizerNames.PAGED_ADAMW 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: 12
training_steps: 430

Training results

Framework versions

PEFT 0.18.1
Transformers 5.5.0
Pytorch 2.8.0+cu128
Datasets 4.5.0
Tokenizers 0.22.1

AI-AgentSafa/ministral-tsql-15

workspace/fine-tuning/outputs/ministral8b-tsql

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

AI-AgentSafa/
ministral-tsql-15