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

You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-babylm-1e-4

like

0

text-generation

generated_from_trainer

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-babylm-1e-4

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

Loss: 2.9816
Accuracy: 0.4424

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
3.5611	1.0	36836	3.4969	0.3818
3.2462	2.0	73672	3.2194	0.4102
3.109	3.0	110508	3.1212	0.4210
3.0408	4.0	147344	3.0694	0.4269
2.9901	5.0	184180	3.0376	0.4308
2.9291	6.0	221016	3.0166	0.4336
2.8924	7.0	257852	3.0016	0.4355
2.861	8.0	294688	2.9941	0.4371
2.8292	9.0	331524	2.9839	0.4384
2.8043	10.0	368360	2.9816	0.4390
2.7779	11.0	405196	2.9767	0.4399
2.7636	12.0	442032	2.9752	0.4404
2.735	13.0	478868	2.9758	0.4408
2.7232	14.0	515704	2.9750	0.4411
2.7068	15.0	552540	2.9745	0.4416
2.6855	16.0	589376	2.9752	0.4418
2.6627	17.0	626212	2.9769	0.4420
2.647	18.0	663048	2.9784	0.4422
2.6294	19.0	699884	2.9807	0.4423
2.6057	20.0	736720	2.9816	0.4424

Framework versions

Transformers 4.35.2
Pytorch 2.1.1+cu121
Datasets 2.15.0
Tokenizers 0.15.0