qwen-text-linear

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: 2.4786

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: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 32
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 20
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
2.4229	0.01	10	2.4927
2.4139	0.03	20	2.4927
2.4284	0.04	30	2.4926
2.4523	0.05	40	2.4923
2.4364	0.06	50	2.4920
2.4522	0.08	60	2.4916
2.5044	0.09	70	2.4910
2.442	0.1	80	2.4903
2.387	0.12	90	2.4895
2.4528	0.13	100	2.4884
2.428	0.14	110	2.4872
2.4669	0.15	120	2.4858
2.4344	0.17	130	2.4842
2.4292	0.18	140	2.4825
2.4199	0.19	150	2.4807
2.3994	0.2	160	2.4790
2.394	0.22	170	2.4771
2.4564	0.23	180	2.4752
2.4278	0.24	190	2.4734
2.4446	0.26	200	2.4717
2.3944	0.27	210	2.4703
2.4077	0.28	220	2.4689
2.4165	0.29	230	2.4678
2.4031	0.31	240	2.4667
2.3953	0.32	250	2.4660
2.3856	0.33	260	2.4651
2.3674	0.35	270	2.4646
2.4261	0.36	280	2.4643
2.4006	0.37	290	2.4638
2.3949	0.38	300	2.4638
2.4207	0.4	310	2.4638
2.4057	0.41	320	2.4638
2.4168	0.42	330	2.4641
2.395	0.44	340	2.4644
2.4121	0.45	350	2.4649
2.4041	0.46	360	2.4655
2.3799	0.47	370	2.4662
2.4049	0.49	380	2.4669
2.3492	0.5	390	2.4676
2.4105	0.51	400	2.4681
2.3855	0.52	410	2.4688
2.4315	0.54	420	2.4694
2.4082	0.55	430	2.4703
2.4911	0.56	440	2.4708
2.4467	0.58	450	2.4716
2.4166	0.59	460	2.4721
2.4174	0.6	470	2.4727
2.392	0.61	480	2.4732
2.37	0.63	490	2.4737
2.3936	0.64	500	2.4742
2.3892	0.65	510	2.4747
2.4089	0.67	520	2.4750
2.4316	0.68	530	2.4753
2.4167	0.69	540	2.4757
2.3993	0.7	550	2.4758
2.4529	0.72	560	2.4761
2.4466	0.73	570	2.4763
2.3683	0.74	580	2.4766
2.4494	0.76	590	2.4768
2.4218	0.77	600	2.4769
2.4086	0.78	610	2.4773
2.4271	0.79	620	2.4774
2.4317	0.81	630	2.4776
2.4495	0.82	640	2.4777
2.4143	0.83	650	2.4779
2.3874	0.84	660	2.4780
2.4044	0.86	670	2.4781
2.3622	0.87	680	2.4781
2.3624	0.88	690	2.4784
2.3998	0.9	700	2.4783
2.3894	0.91	710	2.4784
2.4157	0.92	720	2.4785
2.4197	0.93	730	2.4785
2.3727	0.95	740	2.4786
2.3622	0.96	750	2.4786
2.3938	0.97	760	2.4787
2.404	0.99	770	2.4786
2.4571	1.0	780	2.4786

Framework versions

Transformers 4.35.0.dev0
Pytorch 2.0.1+cu117
Datasets 2.5.2
Tokenizers 0.14.0