gemma-2b-it-lora-math

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

• Loss: 0.4824

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.5108	0.5131	200	0.5543
0.4467	1.0263	400	0.5035
0.4167	1.5394	600	0.4929
0.3823	2.0526	800	0.4840
0.379	2.5657	1000	0.4824

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1