You can deploy this model and start earning money today!

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4

Name: smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4
Availability: InStock

generated_from_trainer

kanishka/counterfactual-babylm-aanns_indef_non_num_removal

model-index

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4

This model was trained from scratch on the kanishka/counterfactual-babylm-aanns_indef_non_num_removal dataset. It achieves the following results on the evaluation set:

Loss: 3.4108
Accuracy: 0.4083

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
3.7334	1.0	18592	3.9107	0.3461
3.4315	2.0	37184	3.6329	0.3754
3.291	3.0	55776	3.5006	0.3893
3.2037	4.0	74368	3.4120	0.3973
3.1447	5.0	92960	3.4094	0.3997
3.0929	6.0	111552	3.3988	0.4018
3.0618	7.0	130144	3.3675	0.4038
3.0253	8.0	148736	3.3760	0.4047
2.9999	9.0	167328	3.3636	0.4058
2.9693	10.0	185920	3.3796	0.4061
2.9531	11.0	204512	3.3582	0.4069
2.9275	12.0	223104	3.3550	0.4077
2.9018	13.0	241696	3.3748	0.4081
2.8867	14.0	260288	3.3719	0.4084
2.8637	15.0	278880	3.3820	0.4083
2.845	16.0	297472	3.3909	0.4084
2.8247	17.0	316064	3.3850	0.4084
2.8071	18.0	334656	3.3954	0.4086
2.7843	19.0	353248	3.4035	0.4086
2.7704	20.0	371840	3.4108	0.4083

Framework versions

Transformers 4.36.0
Pytorch 2.1.1+cu121
Datasets 2.15.0
Tokenizers 0.15.0

kanishka/smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4

smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

kanishka/
smolm-autoreg-bpe-counterfactual-babylm-aann-indef-non_num_removal-3e-4