easylm-rm-gemma-2-9b

This model is a fine-tuned version of scottsuk0306/easylm-sft-gemma-2-9b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6317
• Accuracy: 0.6771

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: 1e-06
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.662	0.0667	10	0.7110	0.5833
0.6752	0.1333	20	0.6839	0.5938
0.682	0.2	30	0.6653	0.6875
0.678	0.2667	40	0.6628	0.7188
0.6591	0.3333	50	0.6403	0.6458
0.6591	0.4	60	0.6619	0.7083
0.6324	0.4667	70	0.6534	0.6875
0.6812	0.5333	80	0.6372	0.6667
0.6562	0.6	90	0.6301	0.6667
0.6534	0.6667	100	0.6283	0.7083
0.6479	0.7333	110	0.6286	0.6875
0.651	0.8	120	0.6281	0.6979
0.6612	0.8667	130	0.6297	0.6979
0.634	0.9333	140	0.6300	0.7083
0.6311	1.0	150	0.6317	0.6771

Framework versions

• Transformers 4.43.3
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1