README.md

metadata

license: mit
tags:
  - generated_from_trainer
model-index:
  - name: gpt2-kl_01_04-hs_cn-loto_lgbt
    results: []

gpt2-kl_01_04-hs_cn-loto_lgbt

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

• Loss: 0.5297

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 4
• seed: 21
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
72.2334	0.03	10	64.5474
30.2905	0.06	20	17.8612
8.886	0.08	30	6.3404
3.4861	0.11	40	2.6744
1.6441	0.14	50	1.1987
0.9746	0.17	60	0.8715
1.0152	0.2	70	0.7307
0.7374	0.23	80	0.6868
0.6436	0.25	90	0.6203
0.7525	0.28	100	0.6001
0.6146	0.31	110	0.5946
0.5676	0.34	120	0.5914
0.563	0.37	130	0.5716
0.6439	0.4	140	0.5743
0.5706	0.42	150	0.5702
0.689	0.45	160	0.5696
0.5986	0.48	170	0.5557
0.6159	0.51	180	0.5606
0.5925	0.54	190	0.5498
0.6124	0.57	200	0.5496
0.559	0.59	210	0.5501
0.6202	0.62	220	0.5544
0.6504	0.65	230	0.5486
0.697	0.68	240	0.5528
0.5171	0.71	250	0.5522
0.6247	0.74	260	0.5390
0.5882	0.76	270	0.5350
0.5941	0.79	280	0.5339
0.5673	0.82	290	0.5321
0.6336	0.85	300	0.5307
0.5581	0.88	310	0.5264
0.5499	0.91	320	0.5251
0.5626	0.93	330	0.5227
0.5443	0.96	340	0.5205
0.6252	0.99	350	0.5215
0.5427	1.02	360	0.5241
0.5231	1.05	370	0.5297

Framework versions

• Transformers 4.28.0
• Pytorch 1.11.0+cu113
• Datasets 2.11.0
• Tokenizers 0.13.3