run_gemma-2-2b_20250507_215200

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

• Loss: 0.3085
• Accuracy: 0.9522
• Precision General: 0.9810
• Recall General: 0.9537
• F1 General: 0.9671
• Precision Memo: 0.8831
• Recall Memo: 0.9444
• F1 Memo: 0.9128
• Precision Album: 0.8333
• Recall Album: 1.0
• F1 Album: 0.9091

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision General	Recall General	F1 General	Precision Memo	Recall Memo	F1 Memo	Precision Album	Recall Album	F1 Album
0.3519	1.0	147	0.2828	0.9144	0.9481	0.9349	0.9415	0.8228	0.9028	0.8609	1.0	0.2	0.3333
0.2311	2.0	294	0.6490	0.8801	0.9592	0.8744	0.9148	0.7363	0.9306	0.8221	0.4	0.4	0.4
0.2362	3.0	441	0.2727	0.9349	0.9375	0.9767	0.9567	0.9254	0.8611	0.8921	1.0	0.2	0.3333
0.1502	4.0	588	0.4032	0.9212	0.9444	0.9488	0.9466	0.8514	0.875	0.8630	1.0	0.4	0.5714
0.0399	5.0	735	0.3743	0.9452	0.9543	0.9721	0.9631	0.9155	0.9028	0.9091	1.0	0.4	0.5714
0.0442	6.0	882	0.5794	0.9418	0.9420	0.9814	0.9613	0.9403	0.875	0.9065	1.0	0.2	0.3333
0.0285	7.0	1029	0.5223	0.9349	0.9455	0.9674	0.9563	0.9	0.875	0.8873	1.0	0.4	0.5714
0.0	8.0	1176	0.6654	0.9384	0.9378	0.9814	0.9591	0.9394	0.8611	0.8986	1.0	0.2	0.3333
0.0048	9.0	1323	0.5900	0.9349	0.9455	0.9674	0.9563	0.9	0.875	0.8873	1.0	0.4	0.5714
0.0014	10.0	1470	0.6443	0.9384	0.9457	0.9721	0.9587	0.9130	0.875	0.8936	1.0	0.4	0.5714
0.0	11.0	1617	0.6786	0.9384	0.9417	0.9767	0.9589	0.9265	0.875	0.9	1.0	0.2	0.3333
0.0	12.0	1764	0.6797	0.9384	0.9417	0.9767	0.9589	0.9265	0.875	0.9	1.0	0.2	0.3333
0.0	13.0	1911	0.6791	0.9384	0.9417	0.9767	0.9589	0.9265	0.875	0.9	1.0	0.2	0.3333
0.0	14.0	2058	0.6785	0.9384	0.9417	0.9767	0.9589	0.9265	0.875	0.9	1.0	0.2	0.3333
0.0	15.0	2205	0.6798	0.9384	0.9417	0.9767	0.9589	0.9265	0.875	0.9	1.0	0.2	0.3333

Framework versions

• PEFT 0.15.0
• Transformers 4.50.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1