synthAIze_telugu_colloquial_trans

This model is a fine-tuned version of unsloth/tinyllama-chat-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:

• Loss: 9.3740

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• num_epochs: 7
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
14.5924	0.0261	4	9.4984
14.7094	0.0521	8	9.4984
14.6701	0.0782	12	9.4984
12.8211	0.1042	16	9.4985
9.5509	0.1303	20	9.4985
8.5853	0.1564	24	9.4986
5.5634	0.1824	28	9.4977
4.6341	0.2085	32	9.4931
4.2825	0.2345	36	9.4830
4.1122	0.2606	40	9.4710
4.1675	0.2866	44	9.4615
4.0636	0.3127	48	9.4545
3.9839	0.3388	52	9.4433
3.9887	0.3648	56	9.4321
3.9356	0.3909	60	9.4303
3.9948	0.4169	64	9.4278
3.9081	0.4430	68	9.4299
3.9618	0.4691	72	9.4265
3.988	0.4951	76	9.4219
4.0317	0.5212	80	9.4220
3.9991	0.5472	84	9.4183
3.8728	0.5733	88	9.4186
4.0057	0.5993	92	9.4189
3.9544	0.6254	96	9.4179
3.9929	0.6515	100	9.4196
3.9485	0.6775	104	9.4146
3.9823	0.7036	108	9.4137
3.9122	0.7296	112	9.4128
4.0261	0.7557	116	9.4095
3.91	0.7818	120	9.4123
4.0387	0.8078	124	9.4124
3.9793	0.8339	128	9.4130
3.8188	0.8599	132	9.4120
3.909	0.8860	136	9.4108
3.9257	0.9121	140	9.4136
3.9342	0.9381	144	9.4145
3.9592	0.9642	148	9.4175
3.9377	0.9902	152	9.4205
3.8464	1.0130	156	9.4188
3.9341	1.0391	160	9.4166
3.9102	1.0651	164	9.4179
3.9182	1.0912	168	9.4212
3.9517	1.1173	172	9.4220
3.9039	1.1433	176	9.4219
3.9169	1.1694	180	9.4230
3.8886	1.1954	184	9.4186
3.944	1.2215	188	9.4140
3.8584	1.2476	192	9.4161
3.8801	1.2736	196	9.4185
3.87	1.2997	200	9.4145
3.8289	1.3257	204	9.4100
3.8811	1.3518	208	9.4077
3.9058	1.3779	212	9.4038
3.8277	1.4039	216	9.4035
3.9422	1.4300	220	9.4021
3.8817	1.4560	224	9.4000
3.8859	1.4821	228	9.4018
3.9168	1.5081	232	9.4073
3.8357	1.5342	236	9.4063
3.906	1.5603	240	9.4013
3.886	1.5863	244	9.4047
3.8844	1.6124	248	9.4063
3.9357	1.6384	252	9.4055
3.8253	1.6645	256	9.4054
3.7222	1.6906	260	9.4068
3.9175	1.7166	264	9.4095
3.9249	1.7427	268	9.4093
3.9101	1.7687	272	9.4074
3.771	1.7948	276	9.4072
3.7613	1.8208	280	9.4064
3.9083	1.8469	284	9.4088
3.7913	1.8730	288	9.4087
3.8821	1.8990	292	9.4053
3.8123	1.9251	296	9.4038
3.8701	1.9511	300	9.4016
3.9222	1.9772	304	9.4022
3.8515	2.0	308	9.4044
3.746	2.0261	312	9.4035
3.7712	2.0521	316	9.4025
3.8676	2.0782	320	9.4002
3.8346	2.1042	324	9.3997
3.7751	2.1303	328	9.4002
3.8502	2.1564	332	9.4001
3.7592	2.1824	336	9.4002
3.8953	2.2085	340	9.3968
3.7921	2.2345	344	9.3946
3.8732	2.2606	348	9.3939
3.7933	2.2866	352	9.3922
3.8393	2.3127	356	9.3874
3.7572	2.3388	360	9.3874
3.856	2.3648	364	9.3885
3.842	2.3909	368	9.3881
3.891	2.4169	372	9.3874
3.7209	2.4430	376	9.3871
3.9075	2.4691	380	9.3875
3.8392	2.4951	384	9.3872
3.8123	2.5212	388	9.3876
3.8181	2.5472	392	9.3898
3.8469	2.5733	396	9.3908
3.9016	2.5993	400	9.3886
3.7321	2.6254	404	9.3852
3.807	2.6515	408	9.3849
3.8461	2.6775	412	9.3868
3.6914	2.7036	416	9.3895
3.8299	2.7296	420	9.3899
3.7647	2.7557	424	9.3909
3.828	2.7818	428	9.3918
3.8529	2.8078	432	9.3928
3.809	2.8339	436	9.3934
3.8223	2.8599	440	9.3936
3.7237	2.8860	444	9.3916
3.8755	2.9121	448	9.3904
3.8432	2.9381	452	9.3901
3.6971	2.9642	456	9.3897
3.8482	2.9902	460	9.3922
3.7642	3.0130	464	9.3940
3.7789	3.0391	468	9.3948
3.6887	3.0651	472	9.3940
3.8469	3.0912	476	9.3933
3.8252	3.1173	480	9.3926
3.8943	3.1433	484	9.3917
3.7709	3.1694	488	9.3928
3.7592	3.1954	492	9.3922
3.6	3.2215	496	9.3901
3.8423	3.2476	500	9.3903
3.7515	3.2736	504	9.3885
3.7935	3.2997	508	9.3876
3.7854	3.3257	512	9.3883
3.771	3.3518	516	9.3910
3.7701	3.3779	520	9.3934
3.9061	3.4039	524	9.3925
3.8134	3.4300	528	9.3894
3.7618	3.4560	532	9.3885
3.7996	3.4821	536	9.3885
3.6992	3.5081	540	9.3861
3.7507	3.5342	544	9.3859
3.738	3.5603	548	9.3844
3.7863	3.5863	552	9.3817
3.7974	3.6124	556	9.3769
3.722	3.6384	560	9.3766
3.8479	3.6645	564	9.3818
3.76	3.6906	568	9.3866
3.805	3.7166	572	9.3876
3.831	3.7427	576	9.3865
3.8149	3.7687	580	9.3854
3.7358	3.7948	584	9.3846
3.7091	3.8208	588	9.3824
3.7404	3.8469	592	9.3805
3.833	3.8730	596	9.3805
3.7432	3.8990	600	9.3810
3.7569	3.9251	604	9.3815
3.8761	3.9511	608	9.3816
3.7284	3.9772	612	9.3806
3.8089	4.0	616	9.3807
3.6199	4.0261	620	9.3816
3.7248	4.0521	624	9.3830
3.6517	4.0782	628	9.3833
3.7501	4.1042	632	9.3809
3.7044	4.1303	636	9.3792
3.7368	4.1564	640	9.3793
3.6834	4.1824	644	9.3791
3.8192	4.2085	648	9.3787
3.7202	4.2345	652	9.3805
3.7935	4.2606	656	9.3826
3.6746	4.2866	660	9.3834
3.7561	4.3127	664	9.3834
3.6912	4.3388	668	9.3837
3.7322	4.3648	672	9.3860
3.6658	4.3909	676	9.3871
3.7631	4.4169	680	9.3871
3.7579	4.4430	684	9.3872
3.793	4.4691	688	9.3866
3.6951	4.4951	692	9.3843
3.7384	4.5212	696	9.3826
3.634	4.5472	700	9.3813
3.7035	4.5733	704	9.3815
3.7729	4.5993	708	9.3821
3.763	4.6254	712	9.3815
3.6896	4.6515	716	9.3813
3.7827	4.6775	720	9.3818
3.7243	4.7036	724	9.3814
3.6214	4.7296	728	9.3809
3.7144	4.7557	732	9.3796
3.8698	4.7818	736	9.3781
3.8257	4.8078	740	9.3772
3.784	4.8339	744	9.3766
3.7899	4.8599	748	9.3775
3.7396	4.8860	752	9.3784
3.7271	4.9121	756	9.3782
3.807	4.9381	760	9.3783
3.7314	4.9642	764	9.3787
3.811	4.9902	768	9.3795
3.7654	5.0130	772	9.3790
3.7322	5.0391	776	9.3787
3.8652	5.0651	780	9.3792
3.6246	5.0912	784	9.3788
3.6057	5.1173	788	9.3783
3.5923	5.1433	792	9.3776
3.6224	5.1694	796	9.3766
3.6847	5.1954	800	9.3756
3.7316	5.2215	804	9.3757
3.597	5.2476	808	9.3761
3.6709	5.2736	812	9.3751
3.5557	5.2997	816	9.3740
3.6442	5.3257	820	9.3727
3.7169	5.3518	824	9.3737
3.6234	5.3779	828	9.3741
3.6838	5.4039	832	9.3742
3.6614	5.4300	836	9.3750
3.7273	5.4560	840	9.3758
3.7639	5.4821	844	9.3758
3.8223	5.5081	848	9.3755
3.7424	5.5342	852	9.3765
3.6588	5.5603	856	9.3764
3.6917	5.5863	860	9.3762
3.7315	5.6124	864	9.3766
3.7366	5.6384	868	9.3764
3.5316	5.6645	872	9.3761
3.6727	5.6906	876	9.3756
3.6911	5.7166	880	9.3752
3.6286	5.7427	884	9.3746
3.7405	5.7687	888	9.3751
3.6738	5.7948	892	9.3759
3.7719	5.8208	896	9.3762
3.6128	5.8469	900	9.3759
3.7748	5.8730	904	9.3759
3.5543	5.8990	908	9.3763
3.6143	5.9251	912	9.3761
3.7183	5.9511	916	9.3765
3.6779	5.9772	920	9.3767
3.7828	6.0	924	9.3764
3.621	6.0261	928	9.3763
3.6744	6.0521	932	9.3768
3.6675	6.0782	936	9.3772
3.6785	6.1042	940	9.3777
3.7176	6.1303	944	9.3785
3.6924	6.1564	948	9.3788
3.5211	6.1824	952	9.3789
3.7995	6.2085	956	9.3789
3.6982	6.2345	960	9.3786
3.7288	6.2606	964	9.3779
3.7139	6.2866	968	9.3770
3.564	6.3127	972	9.3764
3.7078	6.3388	976	9.3760
3.6165	6.3648	980	9.3760
3.5605	6.3909	984	9.3758
3.7073	6.4169	988	9.3757
3.5997	6.4430	992	9.3756
3.444	6.4691	996	9.3756
3.6781	6.4951	1000	9.3757
3.6395	6.5212	1004	9.3759
3.5381	6.5472	1008	9.3761
3.7774	6.5733	1012	9.3757
3.6378	6.5993	1016	9.3752
3.6952	6.6254	1020	9.3749
3.6099	6.6515	1024	9.3746
3.7006	6.6775	1028	9.3743
3.7079	6.7036	1032	9.3743
3.7302	6.7296	1036	9.3743
3.6339	6.7557	1040	9.3745
3.5705	6.7818	1044	9.3744
3.6991	6.8078	1048	9.3742
3.7199	6.8339	1052	9.3740
3.7193	6.8599	1056	9.3739
3.5579	6.8860	1060	9.3739
3.7278	6.9121	1064	9.3739
3.5131	6.9381	1068	9.3740

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0