Gemma-2-2B_task-1_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-1 dataset. It achieves the following results on the evaluation set:

• Loss: 1.0263

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
2.6097	0.8696	5	2.4178
2.3915	1.9130	11	2.0981
1.8784	2.9565	17	1.6329
1.4387	4.0	23	1.2761
1.107	4.8696	28	1.1339
0.972	5.9130	34	1.0639
0.8841	6.9565	40	1.0263
0.7834	8.0	46	1.0309
0.7282	8.8696	51	1.0628
0.5877	9.9130	57	1.1318
0.4988	10.9565	63	1.2335
0.3489	12.0	69	1.5287
0.2429	12.8696	74	1.6523
0.1687	13.9130	80	1.9331

Framework versions

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