README.md

metadata

base_model: Qwen/Qwen-14B
tags:
  - generated_from_trainer
model-index:
  - name: nampdn-ai_tiny-textbooks
    results: []

nampdn-ai_tiny-textbooks

This model is a fine-tuned version of Qwen/Qwen-14B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3572

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.01
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
2.4651	0.02	200	2.3996
2.4335	0.04	400	2.3799
2.3848	0.06	600	2.3746
2.4037	0.08	800	2.3714
2.3985	0.1	1000	2.3693
2.4072	0.12	1200	2.3673
2.4028	0.14	1400	2.3665
2.3748	0.16	1600	2.3643
2.4119	0.18	1800	2.3635
2.4002	0.2	2000	2.3640
2.3865	0.22	2200	2.3635
2.4	0.24	2400	2.3628
2.4096	0.26	2600	2.3625
2.3976	0.28	2800	2.3614
2.3767	0.3	3000	2.3618
2.4151	0.32	3200	2.3616
2.3835	0.34	3400	2.3605
2.3995	0.36	3600	2.3608
2.4121	0.38	3800	2.3602
2.4262	0.4	4000	2.3591
2.3604	0.42	4200	2.3594
2.3954	0.44	4400	2.3594
2.3743	0.46	4600	2.3587
2.4069	0.48	4800	2.3591
2.4103	0.5	5000	2.3585
2.4133	0.52	5200	2.3585
2.4229	0.54	5400	2.3578
2.4397	0.56	5600	2.3581
2.4237	0.58	5800	2.3581
2.4109	0.6	6000	2.3577
2.43	0.62	6200	2.3575
2.3999	0.64	6400	2.3572
2.3771	0.66	6600	2.3577
2.4119	0.68	6800	2.3576
2.3877	0.7	7000	2.3576
2.411	0.72	7200	2.3569
2.3808	0.74	7400	2.3570
2.3989	0.76	7600	2.3571
2.422	0.78	7800	2.3569
2.3768	0.8	8000	2.3569
2.3988	0.82	8200	2.3572
2.3927	0.84	8400	2.3572
2.3961	0.86	8600	2.3573
2.4021	0.88	8800	2.3570
2.3889	0.9	9000	2.3570
2.404	0.92	9200	2.3570
2.3982	0.94	9400	2.3572
2.4018	0.96	9600	2.3573
2.3717	0.98	9800	2.3572
2.4076	1.0	10000	2.3572

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0
• Datasets 2.14.5
• Tokenizers 0.14.1