metadata

license: apache-2.0
base_model: riotu-lab/ArabianGPT-01B
tags:
  - generated_from_trainer
metrics:
  - bleu
  - rouge
model-index:
  - name: res_nw_dj
    results: []

res_nw_dj

This model is a fine-tuned version of riotu-lab/ArabianGPT-01B on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6266
Bleu: 0.2679
Rouge1: 0.5977
Rouge2: 0.3443
Rougel: 0.5959

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Bleu	Validation Loss	Rouge1	Rouge2	Rougel
1.2513	1.0	2703	0.2201	0.7112	0.5172	0.2533	0.5148
0.6462	2.0	5406	0.2451	0.6569	0.5579	0.2974	0.5553
0.5673	3.0	8109	0.2518	0.6499	0.5702	0.3118	0.5677
0.556	4.0	10812	0.2598	0.6330	0.5829	0.3262	0.5807
0.4786	5.0	13515	0.2679	0.6266	0.5977	0.3443	0.5959
0.4123	6.0	16218	0.2707	0.6303	0.6044	0.3548	0.6027
0.3573	7.0	18921	0.2739	0.6372	0.6108	0.3631	0.6088
0.3108	8.0	21624	0.2760	0.6466	0.6130	0.3678	0.6112
0.2719	9.0	24327	0.2802	0.6551	0.6178	0.3727	0.6159
0.2396	10.0	27030	0.2770	0.6658	0.6173	0.3729	0.6153
0.2129	11.0	29733	0.6768	0.2784	0.6188	0.3762	0.6171
0.191	12.0	32436	0.6870	0.2780	0.6208	0.3781	0.6189
0.1733	13.0	35139	0.6967	0.2800	0.6214	0.3799	0.6195
0.1593	14.0	37842	0.7063	0.2790	0.6215	0.3799	0.6195
0.1478	15.0	40545	0.7139	0.2809	0.6216	0.3804	0.6197

Framework versions

Transformers 4.45.0.dev0
Pytorch 2.3.1+cu121
Datasets 2.19.2
Tokenizers 0.19.1