pretrain_wo-cot_w-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: 9.5465

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
11.1865	0.1290	1	11.1164
11.1109	0.2581	2	11.0354
11.0508	0.3871	3	10.9345
10.9774	0.5161	4	10.8247
10.8372	0.6452	5	10.7115
10.742	0.7742	6	10.5983
10.6067	0.9032	7	10.4797
10.5182	1.0323	8	10.3694
10.3704	1.1613	9	10.2614
10.2903	1.2903	10	10.1587
10.1627	1.4194	11	10.0626
10.1078	1.5484	12	9.9744
10.0265	1.6774	13	9.8932
9.9159	1.8065	14	9.8200
9.8432	1.9355	15	9.7532
9.7643	2.0645	16	9.6953
9.7308	2.1935	17	9.6462
9.6923	2.3226	18	9.6069
9.6348	2.4516	19	9.5776
9.5869	2.5806	20	9.5566
9.6044	2.7097	21	9.5465

Framework versions

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