smolm-autoreg-bpe-counterfactual-babylm-indef-anan-1e-3 – 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-3

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-3

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

Loss: 3.4352
Accuracy: 0.4091

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
3.529	1.0	37190	3.7336	0.3656
3.3227	2.0	74380	3.5026	0.3875
3.2211	3.0	111570	3.4307	0.3956
3.1549	4.0	148760	3.4104	0.4000
3.1042	5.0	185950	3.3540	0.4081
3.0649	6.0	223140	3.3679	0.4067
3.0354	7.0	260330	3.3612	0.4067
3.0057	8.0	297520	3.3286	0.4122
2.9771	9.0	334710	3.3623	0.4088
2.9514	10.0	371900	3.3526	0.4091
2.9324	11.0	409090	3.3801	0.4086
2.9125	12.0	446280	3.3819	0.4101
2.8878	13.0	483470	3.3952	0.4081
2.8694	14.0	520660	3.3821	0.4102
2.8447	15.0	557850	3.3905	0.4100
2.8259	16.0	595040	3.3960	0.4102
2.8158	17.0	632230	3.4224	0.4084
2.7923	18.0	669420	3.4164	0.4094
2.7687	19.0	706610	3.4260	0.4093
2.7621	20.0	743800	3.4352	0.4091

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.14.1