age_sentence_conv

This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5164

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: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: reduce_lr_on_plateau
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
4.5887	0.0254	1000	4.3882
4.1742	0.0508	2000	4.1169
4.057	0.0762	3000	3.9806
3.9569	0.1016	4000	3.8949
3.8973	0.1270	5000	3.8564
3.8609	0.1524	6000	3.8018
3.781	0.1778	7000	3.7582
3.7907	0.2032	8000	3.7497
3.7587	0.2286	9000	3.7056
3.6676	0.2540	10000	3.6830
3.7266	0.2794	11000	3.7069
3.6977	0.3048	12000	3.6512
3.6274	0.3302	13000	3.6647
3.6566	0.3555	14000	3.6371
3.6203	0.3809	15000	3.6135
3.6104	0.4063	16000	3.6291
3.6184	0.4317	17000	3.6047
3.5547	0.4571	18000	3.5879
3.6023	0.4825	19000	3.6041
3.5842	0.5079	20000	3.5750
3.5007	0.5333	21000	3.5668
3.5785	0.5587	22000	3.6071
3.5572	0.5841	23000	3.5589
3.5092	0.6095	24000	3.5805
3.5458	0.6349	25000	3.5620
3.5082	0.6603	26000	3.5445
3.5118	0.6857	27000	3.5696
3.5261	0.7111	28000	3.5462
3.4636	0.7365	29000	3.5313
3.5262	0.7619	30000	3.5510
3.5087	0.7873	31000	3.5307
3.4283	0.8127	32000	3.5291
3.5104	0.8381	33000	3.5658
3.4848	0.8635	34000	3.5193
3.4542	0.8889	35000	3.5472
3.4854	0.9143	36000	3.5277
3.4471	0.9397	37000	3.5127
3.4658	0.9651	38000	3.5353
3.4726	0.9905	39000	3.5164

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1