You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-adj_num_freq_balanced-1e-3

Name: smolm-autoreg-bpe-counterfactual-babylm-adj_num_freq_balanced-1e-3
Availability: InStock

generated_from_trainer

kanishka/counterfactual_babylm_aann_excess_adj_removal

model-index

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

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

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

Loss: 3.4904
Accuracy: 0.4056

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: 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
3.5233	1.0	37258	3.7374	0.3658
3.3131	2.0	74516	3.5591	0.3855
3.2155	3.0	111774	3.4638	0.3939
3.1471	4.0	149032	3.4658	0.3964
3.0992	5.0	186290	3.4396	0.3985
3.0583	6.0	223548	3.3943	0.4039
3.0231	7.0	260806	3.3958	0.4046
2.9862	8.0	298064	3.4018	0.4053
2.9604	9.0	335322	3.3988	0.4071
2.935	10.0	372580	3.3970	0.4081
2.9131	11.0	409838	3.3988	0.4076
2.8857	12.0	447096	3.4045	0.4076
2.8613	13.0	484354	3.4254	0.4063
2.8427	14.0	521612	3.4185	0.4071
2.8172	15.0	558870	3.4428	0.4072
2.8031	16.0	596128	3.4478	0.4060
2.7782	17.0	633386	3.4540	0.4064
2.7535	18.0	670644	3.4752	0.4059
2.7371	19.0	707902	3.4821	0.4059
2.7238	20.0	745160	3.4904	0.4056

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-adj_num_freq_balanced-1e-3

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