PaperPrism_gemma2

This model is a fine-tuned version of google/gemma-2-2b-it on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.2055
• Accuracy: 0.7476
• F1: 0.7142
• Precision: 0.7845
• Recall: 0.7476

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	0.0300	100	1.5942	0.7248	0.7061	0.7081	0.7248
No log	0.0601	200	1.6005	0.7524	0.7241	0.7486	0.7524
No log	0.0901	300	1.4531	0.7656	0.7506	0.7555	0.7656
No log	0.1202	400	1.8157	0.7332	0.7134	0.7484	0.7332
0.591	0.1502	500	1.4562	0.7825	0.7753	0.7774	0.7825
0.591	0.1803	600	1.3786	0.7692	0.7636	0.7816	0.7692
0.591	0.2103	700	2.2055	0.7476	0.7142	0.7845	0.7476

Framework versions

• PEFT 0.13.0
• Transformers 4.46.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0