train_boolq_1745950273

This model is a fine-tuned version of google/gemma-3-1b-it on the boolq dataset. It achieves the following results on the evaluation set:

• Loss: 0.1538
• Num Input Tokens Seen: 34633072

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: 2
• eval_batch_size: 2
• seed: 123
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• training_steps: 40000

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.2294	0.0943	200	0.2528	174096
0.4076	0.1886	400	0.2382	344560
0.3531	0.2829	600	0.2216	517536
0.2417	0.3772	800	0.1999	696016
0.1596	0.4715	1000	0.2032	868992
0.2865	0.5658	1200	0.1945	1040544
0.4192	0.6601	1400	0.1821	1211680
0.1268	0.7544	1600	0.1738	1381792
0.172	0.8487	1800	0.1745	1559456
0.3152	0.9430	2000	0.1766	1735840
0.1413	1.0372	2200	0.1739	1910848
0.1775	1.1315	2400	0.2023	2081696
0.1675	1.2258	2600	0.1657	2255952
0.1501	1.3201	2800	0.1832	2427152
0.0747	1.4144	3000	0.1788	2601296
0.0954	1.5087	3200	0.1857	2774672
0.0947	1.6030	3400	0.1752	2944896
0.1546	1.6973	3600	0.1971	3117216
0.2168	1.7916	3800	0.1538	3287952
0.2023	1.8859	4000	0.1737	3464640
0.0789	1.9802	4200	0.1676	3638880
0.0811	2.0745	4400	0.2531	3812624
0.036	2.1688	4600	0.2118	3986544
0.0806	2.2631	4800	0.1946	4158272
0.2564	2.3574	5000	0.2272	4328240
0.0797	2.4517	5200	0.2316	4507760
0.204	2.5460	5400	0.2353	4681664
0.0162	2.6403	5600	0.2062	4856928
0.0736	2.7346	5800	0.2187	5024976
0.1102	2.8289	6000	0.1996	5202368
0.0752	2.9231	6200	0.2483	5377360
0.0575	3.0174	6400	0.2213	5550480
0.0373	3.1117	6600	0.3116	5724080
0.1205	3.2060	6800	0.2736	5896688
0.0456	3.3003	7000	0.2599	6070544
0.0039	3.3946	7200	0.2499	6244624
0.0714	3.4889	7400	0.2836	6416176
0.1576	3.5832	7600	0.2947	6587616
0.1611	3.6775	7800	0.2595	6759696
0.1766	3.7718	8000	0.2564	6932384
0.0044	3.8661	8200	0.2501	7103328
0.0091	3.9604	8400	0.2621	7276304
0.0009	4.0547	8600	0.3677	7448112
0.0241	4.1490	8800	0.3428	7623632
0.0001	4.2433	9000	0.4377	7799248
0.0272	4.3376	9200	0.3371	7974368
0.0004	4.4319	9400	0.4070	8146384
0.0087	4.5262	9600	0.3419	8321456
0.0015	4.6205	9800	0.3727	8490096
0.0006	4.7148	10000	0.3746	8665904
0.0074	4.8091	10200	0.4006	8837712
0.0006	4.9033	10400	0.3620	9010400
0.0738	4.9976	10600	0.3870	9185584
0.0004	5.0919	10800	0.4343	9358160
0.0001	5.1862	11000	0.5670	9535520
0.0	5.2805	11200	0.6110	9709232
0.1252	5.3748	11400	0.6078	9880896
0.0001	5.4691	11600	0.4508	10053056
0.0551	5.5634	11800	0.4355	10229152
0.0003	5.6577	12000	0.4886	10404384
0.0074	5.7520	12200	0.4803	10573872
0.0002	5.8463	12400	0.4446	10748304
0.067	5.9406	12600	0.5511	10917920
0.0001	6.0349	12800	0.5721	11092736
0.0	6.1292	13000	0.5537	11269264
0.0	6.2235	13200	0.5562	11441120
0.0	6.3178	13400	0.7728	11614176
0.0	6.4121	13600	0.5305	11785424
0.0	6.5064	13800	0.6761	11960752
0.0	6.6007	14000	0.5254	12132672
0.0	6.6950	14200	0.5536	12303424
0.0003	6.7893	14400	0.5881	12474592
0.0	6.8835	14600	0.6505	12649424
0.0	6.9778	14800	0.5357	12821280
0.0	7.0721	15000	0.6436	12996208
0.0	7.1664	15200	0.6179	13172592
0.0	7.2607	15400	0.6625	13342864
0.0	7.3550	15600	0.6576	13515600
0.1354	7.4493	15800	0.7310	13688640
0.0002	7.5436	16000	0.5292	13863312
0.0	7.6379	16200	0.6760	14032992
0.0021	7.7322	16400	0.5667	14205936
0.0	7.8265	16600	0.5081	14378336
0.0	7.9208	16800	0.6207	14551456
0.0	8.0151	17000	0.7328	14730672
0.1146	8.1094	17200	0.5845	14904544
0.0	8.2037	17400	0.6697	15078832
0.0	8.2980	17600	0.6067	15254544
0.0001	8.3923	17800	0.5690	15422256
0.0	8.4866	18000	0.7316	15595776
0.0	8.5809	18200	0.6704	15768288
0.1208	8.6752	18400	0.7145	15941776
0.0	8.7694	18600	0.6353	16115152
0.0001	8.8637	18800	0.5448	16284384
0.123	8.9580	19000	0.4909	16457552
0.0001	9.0523	19200	0.4956	16632272
0.0569	9.1466	19400	0.6051	16806304
0.0	9.2409	19600	0.5905	16979072
0.0	9.3352	19800	0.6460	17150160
0.0	9.4295	20000	0.6048	17321280
0.0	9.5238	20200	0.6894	17495488
0.0001	9.6181	20400	0.6447	17670576
0.1459	9.7124	20600	0.6084	17843440
0.0	9.8067	20800	0.5676	18012496
0.0	9.9010	21000	0.6301	18186480
0.0	9.9953	21200	0.7091	18360368
0.0	10.0896	21400	0.7228	18539664
0.0	10.1839	21600	0.5961	18718016
0.0	10.2782	21800	0.6606	18888560
0.0	10.3725	22000	0.6740	19061328
0.0	10.4668	22200	0.6405	19236176
0.0002	10.5611	22400	0.7411	19404288
0.0	10.6554	22600	0.7990	19574224
0.0	10.7496	22800	0.7424	19744496
0.0	10.8439	23000	0.6840	19915984
0.0	10.9382	23200	0.7526	20090944
0.0	11.0325	23400	0.7356	20264992
0.0	11.1268	23600	0.7534	20437952
0.0	11.2211	23800	0.6226	20611040
0.0	11.3154	24000	0.6761	20787488
0.0	11.4097	24200	0.7833	20958240
0.0	11.5040	24400	0.8089	21133392
0.0	11.5983	24600	0.5762	21303360
0.0	11.6926	24800	0.6816	21475184
0.0	11.7869	25000	0.6941	21649744
0.0392	11.8812	25200	0.7063	21819728
0.0	11.9755	25400	0.7234	21993120
0.0	12.0698	25600	0.7154	22164624
0.0001	12.1641	25800	0.6475	22340064
0.0	12.2584	26000	0.5743	22515088
0.0	12.3527	26200	0.6342	22692240
0.0	12.4470	26400	0.6528	22864512
0.0	12.5413	26600	0.6088	23037568
0.0	12.6355	26800	0.6722	23207936
0.0	12.7298	27000	0.6821	23381376
0.0	12.8241	27200	0.7039	23553008
0.0	12.9184	27400	0.7463	23722608
0.0	13.0127	27600	0.7673	23892928
0.0	13.1070	27800	0.7768	24063632
0.0	13.2013	28000	0.7842	24237248
0.0	13.2956	28200	0.7988	24411712
0.0	13.3899	28400	0.8151	24584800
0.0	13.4842	28600	0.8249	24759888
0.0	13.5785	28800	0.8314	24936720
0.0	13.6728	29000	0.8392	25110864
0.0	13.7671	29200	0.8851	25284944
0.0	13.8614	29400	0.8816	25456816
0.0	13.9557	29600	0.9020	25631728
0.0	14.0500	29800	0.9063	25801056
0.0	14.1443	30000	0.9111	25978896
0.0	14.2386	30200	0.9165	26156672
0.0	14.3329	30400	0.9197	26330592
0.0	14.4272	30600	0.9362	26502800
0.0	14.5215	30800	0.9417	26671584
0.0	14.6157	31000	0.9498	26845568
0.0	14.7100	31200	0.9565	27017952
0.0	14.8043	31400	0.9647	27191600
0.0	14.8986	31600	0.9717	27362144
0.0	14.9929	31800	0.9754	27536992
0.0	15.0872	32000	0.9888	27707728
0.0	15.1815	32200	0.9849	27886368
0.0	15.2758	32400	0.9882	28061984
0.0	15.3701	32600	0.9837	28233360
0.0	15.4644	32800	1.0049	28411200
0.0	15.5587	33000	1.0165	28582944
0.0	15.6530	33200	1.0125	28756240
0.0	15.7473	33400	1.0162	28926208
0.0	15.8416	33600	1.0269	29096816
0.0	15.9359	33800	1.0293	29267072
0.0	16.0302	34000	1.0407	29435360
0.0	16.1245	34200	1.0487	29610720
0.0	16.2188	34400	1.0502	29781472
0.0	16.3131	34600	1.0507	29959568
0.0	16.4074	34800	1.0549	30134704
0.0	16.5017	35000	1.0570	30305200
0.0	16.5959	35200	1.0530	30478576
0.0	16.6902	35400	1.0656	30647744
0.0	16.7845	35600	1.0645	30823072
0.0	16.8788	35800	1.0705	30996032
0.0	16.9731	36000	1.0821	31167328
0.0	17.0674	36200	1.0706	31341392
0.0	17.1617	36400	1.0773	31515648
0.0	17.2560	36600	1.0880	31690208
0.0	17.3503	36800	1.0854	31868288
0.0	17.4446	37000	1.0881	32041536
0.0	17.5389	37200	1.0959	32213584
0.0	17.6332	37400	1.0957	32385920
0.0	17.7275	37600	1.1022	32555856
0.0	17.8218	37800	1.0966	32729024
0.0	17.9161	38000	1.0943	32902832
0.0	18.0104	38200	1.1011	33076912
0.0	18.1047	38400	1.1012	33248832
0.0	18.1990	38600	1.1079	33420800
0.0	18.2933	38800	1.1023	33594000
0.0	18.3876	39000	1.1009	33765936
0.0	18.4818	39200	1.1061	33936896
0.0	18.5761	39400	1.1135	34110592
0.0	18.6704	39600	1.1081	34284208
0.0	18.7647	39800	1.1088	34458576
0.0	18.8590	40000	1.1102	34633072

Framework versions

• PEFT 0.15.2.dev0
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1