train_VP_cluster_more_tokens – AI Model by ErikBode | AlphaNeural AI

You can deploy this model and start earning money today!

ErikBode/
train_VP_cluster_more_tokens

like

0

generated_from_trainer

google/gemma-2b

Views

No views yet

Visualize in Weights & Biases

train_VP_cluster_more_tokens

This model is a fine-tuned version of google/gemma-2b on the datasets dataset. It achieves the following results on the evaluation set:

Loss: 3.4458
Rouge1: 0.3759
Rouge2: 0.1110
Rougel: 0.3389
Rougelsum: 0.3388

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: 0.00021
train_batch_size: 20
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 40
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 2
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
3.5428	0.1333	200	3.5027	0.3688	0.1086	0.3321	0.3320
3.3421	0.2667	400	3.4763	0.3751	0.1102	0.3390	0.3388
3.5889	0.4	600	3.4587	0.3765	0.1074	0.3354	0.3355
3.7181	0.5333	800	3.4590	0.3715	0.1086	0.3358	0.3357
3.4013	0.6667	1000	3.4505	0.3757	0.1092	0.3358	0.3355
3.6531	0.8	1200	3.4400	0.3776	0.1097	0.3368	0.3368
3.4296	0.9333	1400	3.4423	0.3744	0.1096	0.3397	0.3396
3.0784	1.0667	1600	3.4559	0.3741	0.1121	0.3378	0.3377
3.4127	1.2	1800	3.4495	0.3729	0.1127	0.3373	0.3371
3.3949	1.3333	2000	3.4483	0.3768	0.1121	0.3395	0.3393
3.2568	1.4667	2200	3.4458	0.3759	0.1110	0.3389	0.3388

Framework versions

PEFT 0.11.1
Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1