metadata

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

res_nw_gulf

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: 0.4779
Bleu: 0.2606
Rouge1: 0.6492
Rouge2: 0.4062
Rougel: 0.6482

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
1.9575	1.0	838	0.1790	0.4805	0.5301	0.2585	0.5287
0.422	2.0	1676	0.2112	0.4327	0.5789	0.3115	0.5778
0.331	3.0	2514	0.2289	0.4185	0.6067	0.3463	0.6056
0.2638	4.0	3352	0.2374	0.4183	0.6213	0.3630	0.6195
0.2131	5.0	4190	0.2423	0.4248	0.6280	0.3719	0.6265
0.1756	6.0	5028	0.2511	0.4323	0.6329	0.3818	0.6312
0.148	7.0	5866	0.2458	0.4369	0.6347	0.3891	0.6327
0.1285	8.0	6704	0.2499	0.4448	0.6380	0.3885	0.6368
0.1152	9.0	7542	0.2539	0.4525	0.6407	0.3977	0.6393
0.105	10.0	8380	0.4590	0.2555	0.6441	0.3998	0.6426
0.0982	11.0	9218	0.4646	0.2580	0.6455	0.4019	0.6445
0.0936	12.0	10056	0.4702	0.2572	0.6457	0.4046	0.6446
0.0899	13.0	10894	0.4737	0.2577	0.6488	0.4053	0.6478
0.0871	14.0	11732	0.4779	0.2606	0.6492	0.4062	0.6482

Framework versions

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