smolm-autoreg-bpe-counterfactual-babylm-adj_num_freq_balanced-seed_211-1e-4 – AI Model by kanishka | AlphaNeural AI

You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-adj_num_freq_balanced-seed_211-1e-4

like

0

text-generation

generated_from_trainer

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual-babylm-adj_num_freq_balanced-seed_211-1e-4

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

Loss: 3.4999
Accuracy: 0.4024

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: 211
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
4.039	1.0	18629	4.2723	0.3087
3.5609	2.0	37258	3.7639	0.3620
3.3837	3.0	55887	3.6041	0.3786
3.2841	4.0	74516	3.5678	0.3857
3.2112	5.0	93145	3.5162	0.3911
3.1608	6.0	111774	3.4931	0.3943
3.1164	7.0	130403	3.4825	0.3960
3.0779	8.0	149032	3.4772	0.3975
3.0424	9.0	167661	3.4669	0.3992
3.0188	10.0	186290	3.4526	0.4004
2.9921	11.0	204919	3.4707	0.4006
2.969	12.0	223548	3.4667	0.4011
2.9473	13.0	242177	3.4682	0.4015
2.9239	14.0	260806	3.4683	0.4019
2.9035	15.0	279435	3.4831	0.4016
2.8845	16.0	298064	3.4649	0.4025
2.8611	17.0	316693	3.4866	0.4022
2.8477	18.0	335322	3.4841	0.4027
2.8308	19.0	353951	3.4953	0.4024
2.8148	20.0	372580	3.4999	0.4024

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.14.1