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: 0.4227

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
3.2369	0.8696	5	3.1359
2.5226	1.9130	11	1.7106
1.1016	2.9565	17	0.7912
0.45	4.0	23	0.5232
0.3492	4.8696	28	0.4661
0.2407	5.9130	34	0.4271
0.2032	6.9565	40	0.4227
0.2575	8.0	46	0.4339
0.1486	8.8696	51	0.4843
0.0961	9.9130	57	0.5847
0.0285	10.9565	63	0.7613
0.0231	12.0	69	0.8747
0.0118	12.8696	74	0.9379
0.0095	13.9130	80	1.0329

Framework versions

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

GaetanMichelet/Gemma-2-2B_task-3_60-samples_config-1

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

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

GaetanMichelet/
Gemma-2-2B_task-3_60-samples_config-1