This model was trained from scratch on the kanishka/counterfactual-babylm-new_regex_aanns_removal dataset. It achieves the following results on the evaluation set:

Loss: 3.4207
Accuracy: 0.4092

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: 0.001
train_batch_size: 32
eval_batch_size: 64
seed: 1024
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 32000
num_epochs: 20.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.5956	1.0	18593	3.7590	0.3599
3.3788	2.0	37186	3.5799	0.3792
3.2581	3.0	55779	3.4859	0.3911
3.1746	4.0	74372	3.4290	0.3981
3.1224	5.0	92965	3.3978	0.4016
3.0799	6.0	111558	3.3948	0.4033
3.0446	7.0	130151	3.3838	0.4053
3.0132	8.0	148744	3.3775	0.4070
2.979	9.0	167337	3.3647	0.4067
2.956	10.0	185930	3.3636	0.4085
2.9367	11.0	204523	3.3768	0.4087
2.9137	12.0	223116	3.3589	0.4092
2.8923	13.0	241709	3.3654	0.4094
2.8716	14.0	260302	3.3621	0.4099
2.8471	15.0	278895	3.4012	0.4089
2.8295	16.0	297488	3.3922	0.4096
2.8136	17.0	316081	3.3921	0.4099
2.7942	18.0	334674	3.4023	0.4094
2.7734	19.0	353267	3.4129	0.4094
2.758	20.0	371860	3.4207	0.4092

Framework versions

Transformers 4.40.1
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.19.1

kanishka/smolm-autoreg-bpe-counterfactual-babylm-new_regex_aanns_removal-seed_1024-1e-3

smolm-autoreg-bpe-counterfactual-babylm-new_regex_aanns_removal-1e-3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-new_regex_aanns_removal-seed_1024-1e-3