SmolLM2-FT-MyDataset

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-135M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5380

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: 5e-05
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 3
• total_train_batch_size: 3
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
1.3859	1.1538	50	1.8652
1.2355	2.3077	100	1.9235
1.1499	3.4615	150	1.9949
1.0908	4.6154	200	2.0660
0.9575	5.7692	250	2.1495
0.8631	6.9231	300	2.2484
0.7775	8.0769	350	2.3646
0.6162	9.2308	400	2.5319
0.5879	10.3846	450	2.6362
0.5224	11.5385	500	2.7548
0.4339	12.6923	550	2.8590
0.4305	13.8462	600	2.9796
0.3826	15.0	650	3.0848
0.3383	16.1538	700	3.2293
0.2905	17.3077	750	3.3046
0.2581	18.4615	800	3.3902
0.2367	19.6154	850	3.4507
0.2223	20.7692	900	3.4877
0.2136	21.9231	950	3.5117
0.217	23.0769	1000	3.5380

Framework versions

• Transformers 4.44.2
• Pytorch 2.8.0+cu126
• Datasets 4.2.0
• Tokenizers 0.19.1