Update README.md

README.md

@@ -12,22 +12,22 @@ tags:
 base_model: EleutherAI/polyglot-ko-12.8b
 ---
 
-This model is a instruct-tuned poylglot-ko-12.8b model, using 10% [Kullm, OIG, KoAlpaca] Instruction dataset.
+This model is a instruct-tuned poylglot-ko-12.8b model, using 10% [Kullm, OIG, KoAlpaca] Instruction dataset. -> 29step
 
 ## Training hyperparameters
 - learning_rate: 5e-5
 - seed: 42
-- distributed_type: multi-GPU (A100 80G)
+- distributed_type: multi-GPU (A100 40G) + CPU offloading (512GB)
+- num_devices: 1
 - train_batch_size: 4
-- num_devices: 4
-- gradient_accumulation_steps: 4
+- gradient_accumulation_steps: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 2.0
 
 ## Framework versions
 - Transformers 4.35.0
-- Pytorch 2.0.1+cu118
+- Pytorch 2.0.1+cu117
 - Datasets 2.14.6
 - deepspeed 0.11.1
 - accelerate 0.24.1