age_transcript_conv1

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

• Loss: 3.2240

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
4.7007	0.0254	1000	4.4922
4.2819	0.0508	2000	4.2123
4.1092	0.0762	3000	4.0324
3.9861	0.1016	4000	3.9267
3.9221	0.1270	5000	3.8582
3.8904	0.1524	6000	3.7809
3.7526	0.1778	7000	3.7252
3.7724	0.2032	8000	3.6846
3.6967	0.2285	9000	3.6293
3.5701	0.2539	10000	3.5902
3.676	0.2793	11000	3.5787
3.6092	0.3047	12000	3.5333
3.5105	0.3301	13000	3.5061
3.5298	0.3555	14000	3.4776
3.4857	0.3809	15000	3.4537
3.4688	0.4063	16000	3.4490
3.4914	0.4317	17000	3.4141
3.3866	0.4571	18000	3.3970
3.484	0.4825	19000	3.3963
3.4187	0.5079	20000	3.3733
3.2706	0.5333	21000	3.3546
3.4344	0.5587	22000	3.3640
3.3577	0.5841	23000	3.3337
3.3058	0.6095	24000	3.3364
3.3558	0.6349	25000	3.3195
3.2865	0.6603	26000	3.2988
3.3295	0.6856	27000	3.3054
3.3024	0.7110	28000	3.2867
3.1984	0.7364	29000	3.2751
3.3467	0.7618	30000	3.2792
3.3066	0.7872	31000	3.2647
3.1441	0.8126	32000	3.2606
3.3292	0.8380	33000	3.2656
3.2561	0.8634	34000	3.2444
3.2027	0.8888	35000	3.2549
3.264	0.9142	36000	3.2401
3.1928	0.9396	37000	3.2293
3.2385	0.9650	38000	3.2376
3.2274	0.9904	39000	3.2240

Framework versions

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