You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-indef-naan-rerun-1e-4

Name: smolm-autoreg-bpe-counterfactual-babylm-indef-naan-rerun-1e-4
Availability: InStock

generated_from_trainer

kanishka/counterfactual_babylm_aann_indef_naan

model-index

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual-babylm-indef-naan-rerun-1e-4

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

Loss: 3.4083
Accuracy: 0.4074

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: 32
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
4.0503	1.0	18595	4.2445	0.3095
3.5628	2.0	37190	3.7291	0.3632
3.387	3.0	55785	3.5789	0.3803
3.2888	4.0	74380	3.5034	0.3888
3.2233	5.0	92975	3.4553	0.3932
3.169	6.0	111570	3.4258	0.3974
3.1266	7.0	130165	3.4167	0.3996
3.0969	8.0	148760	3.4030	0.4014
3.0645	9.0	167355	3.4076	0.4025
3.037	10.0	185950	3.3951	0.4039
3.0085	11.0	204545	3.3910	0.4048
2.9889	12.0	223140	3.3887	0.4054
2.9703	13.0	241735	3.3882	0.4060
2.9498	14.0	260330	3.3978	0.4059
2.9248	15.0	278925	3.3988	0.4062
2.9106	16.0	297520	3.3962	0.4067
2.8909	17.0	316115	3.3952	0.4070
2.8717	18.0	334710	3.3977	0.4072
2.8605	19.0	353305	3.4046	0.4074
2.8433	20.0	371900	3.4083	0.4074

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-indef-naan-rerun-1e-4

smolm-autoreg-bpe-counterfactual-babylm-indef-naan-rerun-1e-4

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-indef-naan-rerun-1e-4