Gemma-2-2B_task-3_60-samples_config-1_full

This model is a fine-tuned version of google/gemma-2-2b-it on the GaetanMichelet/chat-60_ft_task-3 dataset. It achieves the following results on the evaluation set:

• Loss: 0.9621

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: 8
• total_train_batch_size: 8
• 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
1.3556	0.8696	5	1.3616
1.3617	1.9130	11	1.2836
1.257	2.9565	17	1.1817
1.1091	4.0	23	1.0976
0.9986	4.8696	28	1.0301
0.9277	5.9130	34	0.9902
0.8595	6.9565	40	0.9712
0.8452	8.0	46	0.9633
0.8318	8.8696	51	0.9621
0.8165	9.9130	57	0.9639
0.7226	10.9565	63	0.9804
0.6818	12.0	69	1.0013
0.6785	12.8696	74	1.0242
0.5736	13.9130	80	1.0798
0.5264	14.9565	86	1.1451
0.5243	16.0	92	1.2084

Framework versions

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