gemma-2b-dolly-qa

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

• Loss: 2.0223

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• training_steps: 1480

Training results

Training Loss	Epoch	Step	Validation Loss
2.9293	1.6393	100	2.5752
2.4386	3.2787	200	2.2894
2.2528	4.9180	300	2.1724
2.1653	6.5574	400	2.1080
2.1144	8.1967	500	2.0766
2.087	9.8361	600	2.0583
2.0697	11.4754	700	2.0473
2.0493	13.1148	800	2.0395
2.0472	14.7541	900	2.0341
2.0311	16.3934	1000	2.0300
2.029	18.0328	1100	2.0267
2.0233	19.6721	1200	2.0245
2.0177	21.3115	1300	2.0230
2.0136	22.9508	1400	2.0223

Framework versions

• PEFT 0.10.0
• Transformers 4.40.1
• Pytorch 2.1.0.post0+cxx11.abi
• Datasets 2.19.0
• Tokenizers 0.19.1