smolm-autoreg-bpe-counterfactual-babylm-indef-anan-1e-4 – AI Model by kanishka | AlphaNeural AI

You can deploy this model and start earning money today!

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

like

0

text-generation

generated_from_trainer

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual-babylm-indef-anan-1e-4

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.4448
Accuracy: 0.4067

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: 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
3.7101	1.0	37190	3.8901	0.3473
3.4207	2.0	74380	3.6009	0.3779
3.2969	3.0	111570	3.5215	0.3879
3.2188	4.0	148760	3.4783	0.3938
3.1613	5.0	185950	3.4380	0.3973
3.1166	6.0	223140	3.4355	0.3997
3.0841	7.0	260330	3.4268	0.4006
3.0521	8.0	297520	3.3898	0.4037
3.0221	9.0	334710	3.4219	0.4038
2.9956	10.0	371900	3.4051	0.4044
2.9757	11.0	409090	3.4273	0.4043
2.9561	12.0	446280	3.4057	0.4056
2.9323	13.0	483470	3.4212	0.4060
2.9141	14.0	520660	3.4264	0.4063
2.8904	15.0	557850	3.4196	0.4060
2.871	16.0	595040	3.4227	0.4065
2.8644	17.0	632230	3.4368	0.4064
2.8424	18.0	669420	3.4319	0.4067
2.8199	19.0	706610	3.4396	0.4068
2.8162	20.0	743800	3.4448	0.4067

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.14.1