pretrain_w-cot_wo-asd

This model is a fine-tuned version of EleutherAI/polyglot-ko-1.3b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2648

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6934	0.1727	1000	0.2945
0.2864	0.3454	2000	0.2812
0.2787	0.5181	3000	0.2762
0.2739	0.6908	4000	0.2747
0.2709	0.8636	5000	0.2722
0.2727	1.0363	6000	0.2704
0.2694	1.2090	7000	0.2703
0.2684	1.3817	8000	0.2683
0.2652	1.5544	9000	0.2678
0.2641	1.7271	10000	0.2674
0.2624	1.8998	11000	0.2670
0.268	2.0725	12000	0.2661
0.2614	2.2453	13000	0.2661
0.2622	2.4180	14000	0.2656
0.2621	2.5907	15000	0.2653
0.263	2.7634	16000	0.2649
0.2625	2.9361	17000	0.2648

Framework versions

• PEFT 0.11.1
• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1