README.md

metadata

library_name: peft
license: gemma
base_model: google/gemma-2-2b
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: run_gemma-2-2b_20250507_202421
    results: []

run_gemma-2-2b_20250507_202421

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: 0.5589
• Accuracy: 0.7986
• Precision General: 0.7985
• Recall General: 0.9722
• F1 General: 0.8768
• Precision Memo: 0.8
• Recall Memo: 0.3333
• F1 Memo: 0.4706
• Precision Album: 0.0
• Recall Album: 0.0
• F1 Album: 0.0

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
0.8016	1.0	147	1.0285	0.5274	0.816	0.4744	0.6	0.3114	0.7222	0.4351	0.0	0.0	0.0
0.5683	2.0	294	0.6889	0.75	0.8025	0.8884	0.8433	0.5185	0.3889	0.4444	0.0	0.0	0.0
0.5543	3.0	441	0.5514	0.7705	0.7891	0.9395	0.8577	0.6389	0.3194	0.4259	0.0	0.0	0.0
0.4472	4.0	588	0.5336	0.8151	0.8132	0.9721	0.8856	0.8286	0.4028	0.5421	0.0	0.0	0.0
0.4579	5.0	735	0.6379	0.75	0.8114	0.8605	0.8352	0.5312	0.4722	0.5	0.0	0.0	0.0
0.5291	6.0	882	0.6045	0.8048	0.8062	0.9674	0.8795	0.7941	0.375	0.5094	0.0	0.0	0.0
0.4397	7.0	1029	0.5941	0.8116	0.8293	0.9488	0.8850	0.7174	0.4583	0.5593	0.0	0.0	0.0
0.3478	8.0	1176	0.6598	0.8151	0.8061	0.9860	0.8870	0.8966	0.3611	0.5149	0.0	0.0	0.0
0.4765	9.0	1323	0.6486	0.8082	0.8069	0.9721	0.8819	0.8182	0.375	0.5143	0.0	0.0	0.0
0.3421	10.0	1470	0.6713	0.8082	0.8	0.9860	0.8833	0.8889	0.3333	0.4848	0.0	0.0	0.0
0.304	11.0	1617	0.6890	0.8048	0.8038	0.9721	0.88	0.8125	0.3611	0.5	0.0	0.0	0.0
0.3041	12.0	1764	0.6821	0.8014	0.8054	0.9628	0.8771	0.7714	0.375	0.5047	0.0	0.0	0.0
0.3565	13.0	1911	0.6882	0.8048	0.8086	0.9628	0.8790	0.7778	0.3889	0.5185	0.0	0.0	0.0
0.3987	14.0	2058	0.6888	0.8014	0.8054	0.9628	0.8771	0.7714	0.375	0.5047	0.0	0.0	0.0
0.338	15.0	2205	0.6909	0.8014	0.8054	0.9628	0.8771	0.7714	0.375	0.5047	0.0	0.0	0.0

Framework versions

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