smolm-autoreg-bpe-counterfactual-babylm-pipps_and_keys_to_it_all_removal-3e-4 – AI Model by kanishka | AlphaNeural AI

You can deploy this model and start earning money today!

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

like

0

text-generation

generated_from_trainer

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

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

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

Loss: 3.4222
Accuracy: 0.4082

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: 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
3.7348	1.0	18594	3.9020	0.3469
3.4377	2.0	37188	3.6359	0.3749
3.295	3.0	55782	3.5103	0.3884
3.2047	4.0	74376	3.4410	0.3952
3.1471	5.0	92970	3.4172	0.3988
3.101	6.0	111564	3.4061	0.4010
3.0626	7.0	130158	3.3768	0.4034
3.0277	8.0	148752	3.3994	0.4043
3.0062	9.0	167346	3.3710	0.4059
2.9755	10.0	185940	3.3847	0.4059
2.9514	11.0	204534	3.3753	0.4078
2.9304	12.0	223128	3.3837	0.4072
2.9068	13.0	241722	3.3885	0.4079
2.8872	14.0	260316	3.3989	0.4073
2.8636	15.0	278910	3.3993	0.4076
2.8469	16.0	297504	3.3945	0.4081
2.825	17.0	316098	3.4017	0.4083
2.8104	18.0	334692	3.4111	0.4082
2.7905	19.0	353286	3.4110	0.4085
2.7755	20.0	371880	3.4222	0.4082

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.14.1