smolm-autoreg-bpe-counterfactual_babylm_anans_new-1e-3 – AI Model by kanishka | AlphaNeural AI

You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual_babylm_anans_new-1e-3

like

0

text-generation

generated_from_trainer

kanishka/counterfactual_babylm_anans_new

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual_babylm_anans_new-1e-3

This model was trained from scratch on the kanishka/counterfactual_babylm_anans_new dataset. It achieves the following results on the evaluation set:

Loss: 3.4254
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: 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
3.6007	1.0	18595	3.7808	0.3583
3.3831	2.0	37190	3.5820	0.3801
3.2601	3.0	55785	3.4779	0.3914
3.1745	4.0	74380	3.4456	0.3976
3.1238	5.0	92975	3.3903	0.4007
3.081	6.0	111570	3.3846	0.4037
3.0438	7.0	130165	3.3775	0.4049
3.0139	8.0	148760	3.3804	0.4060
2.9849	9.0	167355	3.3752	0.4065
2.9642	10.0	185950	3.3811	0.4078
2.935	11.0	204545	3.3705	0.4076
2.9128	12.0	223140	3.3703	0.4087
2.8963	13.0	241735	3.3833	0.4084
2.8702	14.0	260330	3.3925	0.4090
2.8516	15.0	278925	3.3907	0.4092
2.8317	16.0	297520	3.3914	0.4094
2.8121	17.0	316115	3.4075	0.4087
2.7963	18.0	334710	3.4124	0.4091
2.7801	19.0	353305	3.4181	0.4091
2.7649	20.0	371900	3.4254	0.4091

Framework versions

Transformers 4.40.1
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.19.1