Mistral-7B_task-1_180-samples_config-2_full_auto

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60_ft_task-1_auto, the GaetanMichelet/chat-120_ft_task-1_auto and the GaetanMichelet/chat-180_ft_task-1_auto datasets. It achieves the following results on the evaluation set:

Loss: 0.8198

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
1.6633	0.9412	8	1.5707
1.2689	2.0	17	1.1262
0.9431	2.9412	25	0.8984
0.7827	4.0	34	0.8429
0.7185	4.9412	42	0.8198
0.6317	6.0	51	0.8236
0.4646	6.9412	59	0.8859
0.3485	8.0	68	0.9817
0.2504	8.9412	76	1.1165
0.1848	10.0	85	1.1629
0.1505	10.9412	93	1.2382
0.1227	12.0	102	1.2855