metadata

license: mit
base_model: DTAI-KULeuven/robbert-2023-dutch-large
tags:
  - generated_from_trainer
model-index:
  - name: robbert-2023-dutch-large-ft-lcn
    results: []

robbert-2023-dutch-large-ft-lcn

This model is a fine-tuned version of DTAI-KULeuven/robbert-2023-dutch-large on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.0046

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: 0.0001
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss
2.4819	1.0	69	2.1984
2.1774	2.0	138	2.1685
2.0555	3.0	207	2.1070
1.9779	4.0	276	2.0563
1.9261	5.0	345	2.0365
1.8381	6.0	414	2.0416
1.8136	7.0	483	2.0397
1.7682	8.0	552	2.0639
1.7484	9.0	621	2.0264
1.6742	10.0	690	2.0665
1.6311	11.0	759	2.0448
1.5907	12.0	828	2.0722
1.5301	13.0	897	1.9631
1.5052	14.0	966	2.0467
1.4834	15.0	1035	1.9810
1.4219	16.0	1104	2.0255
1.4029	17.0	1173	2.0746
1.3628	18.0	1242	1.9811
1.3356	19.0	1311	2.0329
1.3028	20.0	1380	2.0039
1.2955	21.0	1449	1.9837
1.2231	22.0	1518	1.9871
1.2093	23.0	1587	2.0143
1.1945	24.0	1656	1.9659
1.1657	25.0	1725	2.0569
1.1369	26.0	1794	1.9878
1.0946	27.0	1863	2.0062
1.063	28.0	1932	2.0421
1.0521	29.0	2001	2.0320
1.0443	30.0	2070	2.0580
1.0325	31.0	2139	1.9606
0.9804	32.0	2208	2.1121
0.9674	33.0	2277	2.0156
0.9563	34.0	2346	2.0292
0.927	35.0	2415	2.0528
0.9236	36.0	2484	1.9851
0.9319	37.0	2553	2.0392
0.8921	38.0	2622	2.0334
0.8742	39.0	2691	2.0492
0.8955	40.0	2760	1.9491

Framework versions

Transformers 4.31.0
Pytorch 1.11.0+cu113
Datasets 2.14.4
Tokenizers 0.13.3