This model is a instruct-tuned poylglot-ko-1.3b model, using only 1% of [Kullm, OIG, KoAlpaca] Instruction dataset. len10_k100_mppl_n0.1.json
Training hyperparameters
- learning_rate: 5e-5
- train_batch_size: 1
- seed: 42
- distributed_type: multi-GPU (A30 24G) + No Offloading
- num_devices: 2
- gradient_accumulation_steps: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1.0
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.11.0
- deepspeed 0.9.5
- Downloads last month
- 1,558
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.