You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4

Name: smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4
Availability: InStock

generated_from_trainer

kanishka/counterfactual_babylm_aann_all_det_removal

model-index

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4

This model was trained from scratch on the kanishka/counterfactual_babylm_aann_all_det_removal dataset. It achieves the following results on the evaluation set:

Loss: 3.4401
Accuracy: 0.4087

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: 0.0003
train_batch_size: 16
eval_batch_size: 64
seed: 42
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
3.5638	1.0	37189	3.7368	0.3622
3.3412	2.0	74378	3.5511	0.3852
3.2337	3.0	111567	3.4790	0.3939
3.1708	4.0	148756	3.4317	0.3994
3.1125	5.0	185945	3.4199	0.4013
3.0765	6.0	223134	3.4033	0.4038
3.0361	7.0	260323	3.3774	0.4055
3.0108	8.0	297512	3.3853	0.4064
2.9931	9.0	334701	3.3719	0.4072
2.9628	10.0	371890	3.3715	0.4079
2.9363	11.0	409079	3.3925	0.4082
2.9167	12.0	446268	3.3869	0.4083
2.8918	13.0	483457	3.3873	0.4089
2.8737	14.0	520646	3.3924	0.4086
2.8545	15.0	557835	3.3917	0.4090
2.8353	16.0	595024	3.4101	0.4089
2.8209	17.0	632213	3.4230	0.4089
2.7977	18.0	669402	3.4256	0.4089
2.781	19.0	706591	3.4295	0.4089
2.7692	20.0	743780	3.4401	0.4087

Framework versions

Transformers 4.36.0
Pytorch 2.1.1+cu121
Datasets 2.15.0
Tokenizers 0.15.0

kanishka/smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4

smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-all_det_removal-3e-4