Gemma-2-2B_task-1_120-samples_config-2_auto

This model is a fine-tuned version of google/gemma-2-2b-it on the GaetanMichelet/chat-60_ft_task-1_auto and the GaetanMichelet/chat-120_ft_task-1_auto datasets. It achieves the following results on the evaluation set:

• Loss: 0.9704

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: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
1.5728	0.9091	5	1.6556
1.5544	2.0	11	1.3106
1.1135	2.9091	16	1.1138
0.9739	4.0	22	1.0347
0.8888	4.9091	27	0.9919
0.8107	6.0	33	0.9704
0.6431	6.9091	38	0.9726
0.5564	8.0	44	1.0095
0.4886	8.9091	49	1.1149
0.3412	10.0	55	1.3270
0.2372	10.9091	60	1.5941
0.1438	12.0	66	1.8735
0.0894	12.9091	71	2.1113

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.1.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1