You can deploy this model and start earning money today!

kanishka/
opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3

Name: opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3
Availability: InStock

generated_from_trainer

kanishka/babylm2-rewritten-clean-spacy_ablate_both_lenient

model-index

autotrain_compatible

text-generation-inference

endpoints_compatible

Views

No views yet

opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3

This model was trained from scratch on the kanishka/babylm2-rewritten-clean-spacy_ablate_both_lenient dataset. It achieves the following results on the evaluation set:

Loss: 2.7016
Accuracy: 0.4768

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
gradient_accumulation_steps: 8
total_train_batch_size: 256
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
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
4.0452	1.0	2161	3.8461	0.3582
3.4044	2.0	4322	3.3373	0.4059
3.0883	3.0	6483	3.1175	0.4280
2.9356	4.0	8644	3.0102	0.4385
2.8412	5.0	10805	2.9480	0.4448
2.7723	6.0	12966	2.9066	0.4491
2.724	7.0	15127	2.8816	0.4522
2.6921	8.0	17288	2.8555	0.4550
2.6667	9.0	19449	2.8422	0.4564
2.6444	10.0	21610	2.8279	0.4580
2.6266	11.0	23771	2.8237	0.4585
2.6091	12.0	25932	2.8160	0.4593
2.6148	13.0	28093	2.8079	0.4605
2.6098	14.0	30254	2.8102	0.4605
2.5997	15.0	32415	2.7975	0.4619
2.5667	16.0	34576	2.7687	0.4655
2.5125	17.0	36737	2.7424	0.4691
2.4545	18.0	38898	2.7210	0.4723
2.3864	19.0	41059	2.7050	0.4751
2.3116	19.9911	43200	2.7016	0.4768

Framework versions

Transformers 4.48.0
Pytorch 2.6.0+cu124
Datasets 3.2.0
Tokenizers 0.21.1

kanishka/opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3

opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

kanishka/
opt-babylm2-rewritten-clean-spacy-earlystop_ablate_both_lenient-bpe_seed-42_1e-3