train_boolq_1745950281

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the boolq dataset. It achieves the following results on the evaluation set:

• Loss: 0.1565
• Num Input Tokens Seen: 37097424

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
0.142	0.0943	200	0.3041	186768
0.1731	0.1886	400	0.2873	369808
0.2001	0.2829	600	0.2180	554928
0.1482	0.3772	800	0.1782	746560
0.0599	0.4715	1000	0.2205	932848
0.3125	0.5658	1200	0.1862	1116128
0.1529	0.6601	1400	0.1673	1299664
0.0853	0.7544	1600	0.1565	1481856
0.2545	0.8487	1800	0.1709	1672160
0.3019	0.9430	2000	0.1679	1860608
0.0054	1.0372	2200	0.1949	2047984
0.0968	1.1315	2400	0.2926	2230960
0.3198	1.2258	2600	0.2244	2417664
0.1708	1.3201	2800	0.2037	2600368
0.002	1.4144	3000	0.2372	2786848
0.1055	1.5087	3200	0.1571	2972672
0.0032	1.6030	3400	0.1902	3154640
0.112	1.6973	3600	0.1928	3339328
0.1738	1.7916	3800	0.1873	3522384
0.2048	1.8859	4000	0.2202	3712352
0.0116	1.9802	4200	0.2042	3899328
0.0833	2.0745	4400	0.2827	4085888
0.0003	2.1688	4600	0.3265	4271936
0.0049	2.2631	4800	0.3109	4456320
0.169	2.3574	5000	0.3515	4638512
0.0012	2.4517	5200	0.2239	4830688
0.0038	2.5460	5400	0.2723	5016480
0.0008	2.6403	5600	0.2932	5204048
0.0006	2.7346	5800	0.3025	5383984
0.001	2.8289	6000	0.3317	5574016
0.0024	2.9231	6200	0.2807	5761616
0.0007	3.0174	6400	0.2992	5948128
0.0001	3.1117	6600	0.3997	6134304
0.0017	3.2060	6800	0.3824	6319616
0.0	3.3003	7000	0.4372	6505744
0.0001	3.3946	7200	0.4133	6692208
0.0002	3.4889	7400	0.4023	6875616
0.0001	3.5832	7600	0.3824	7059472
0.0	3.6775	7800	0.4565	7243472
0.0009	3.7718	8000	0.5463	7428048
0.001	3.8661	8200	0.2695	7611184
0.0015	3.9604	8400	0.3342	7796112
0.1145	4.0547	8600	0.3552	7979520
0.0	4.1490	8800	0.3756	8167776
0.0002	4.2433	9000	0.4579	8355856
0.0	4.3376	9200	0.4628	8543120
0.0002	4.4319	9400	0.4611	8727088
0.0683	4.5262	9600	0.3791	8914992
0.0001	4.6205	9800	0.3732	9095040
0.0001	4.7148	10000	0.3789	9283072
0.0004	4.8091	10200	0.2978	9467600
0.0001	4.9033	10400	0.3437	9653456
0.0009	4.9976	10600	0.3521	9841232
0.2438	5.0919	10800	0.5284	10025504
0.1563	5.1862	11000	0.4800	10216464
0.0012	5.2805	11200	0.4108	10402448
0.0	5.3748	11400	0.4343	10586976
0.0	5.4691	11600	0.4661	10770896
0.0001	5.5634	11800	0.4765	10959424
0.1315	5.6577	12000	0.3611	11146816
0.0	5.7520	12200	0.5469	11328528
0.0	5.8463	12400	0.4576	11515600
0.0	5.9406	12600	0.5431	11697056
0.0	6.0349	12800	0.4700	11884336
0.0	6.1292	13000	0.3942	12074128
0.0	6.2235	13200	0.4035	12258064
0.1252	6.3178	13400	0.5062	12443248
0.0001	6.4121	13600	0.5219	12626480
0.0001	6.5064	13800	0.3889	12813808
0.0	6.6007	14000	0.4314	12998256
0.0	6.6950	14200	0.4322	13180928
0.0	6.7893	14400	0.3947	13364368
0.2907	6.8835	14600	0.3949	13552272
0.0002	6.9778	14800	0.3337	13735904
0.0001	7.0721	15000	0.4086	13924000
0.0005	7.1664	15200	0.3822	14113184
0.0002	7.2607	15400	0.3601	14295568
0.0001	7.3550	15600	0.4522	14480560
0.0	7.4493	15800	0.4758	14664736
0.0	7.5436	16000	0.4593	14852128
0.0	7.6379	16200	0.4756	15033840
0.0001	7.7322	16400	0.4497	15219136
0.0001	7.8265	16600	0.4442	15404160
0.0001	7.9208	16800	0.3951	15589632
0.0	8.0151	17000	0.4466	15781760
0.1486	8.1094	17200	0.4521	15967648
0.0001	8.2037	17400	0.5410	16155248
0.0	8.2980	17600	0.5341	16343648
0.0	8.3923	17800	0.4714	16523360
0.0001	8.4866	18000	0.4917	16709008
0.0	8.5809	18200	0.5282	16893648
0.0	8.6752	18400	0.4734	17079824
0.0002	8.7694	18600	0.3915	17265072
0.0001	8.8637	18800	0.4695	17445904
0.0001	8.9580	19000	0.4629	17631504
0.0	9.0523	19200	0.4747	17818512
0.0	9.1466	19400	0.4857	18005200
0.0	9.2409	19600	0.5190	18190416
0.0	9.3352	19800	0.5195	18373200
0.0	9.4295	20000	0.4981	18556672
0.0	9.5238	20200	0.4594	18742816
0.0	9.6181	20400	0.5346	18930224
0.0001	9.7124	20600	0.4166	19115456
0.1314	9.8067	20800	0.4289	19296016
0.0	9.9010	21000	0.4743	19482416
0.0	9.9953	21200	0.5044	19668640
0.0	10.0896	21400	0.5451	19860880
0.0	10.1839	21600	0.4978	20052672
0.0	10.2782	21800	0.5213	20236224
0.0	10.3725	22000	0.5452	20421632
0.0	10.4668	22200	0.5131	20608320
0.0	10.5611	22400	0.5357	20788112
0.0001	10.6554	22600	0.4131	20969744
0.0	10.7496	22800	0.5415	21151648
0.0	10.8439	23000	0.4808	21335600
0.0	10.9382	23200	0.5149	21522352
0.0	11.0325	23400	0.5375	21709568
0.0	11.1268	23600	0.5517	21894592
0.0	11.2211	23800	0.5750	22079344
0.0	11.3154	24000	0.5305	22269152
0.0	11.4097	24200	0.5367	22451760
0.0	11.5040	24400	0.5912	22639312
0.0	11.5983	24600	0.5122	22821728
0.0	11.6926	24800	0.5314	23005696
0.0	11.7869	25000	0.4149	23192112
0.0001	11.8812	25200	0.4076	23373840
0.0001	11.9755	25400	0.3990	23559968
0.0	12.0698	25600	0.4313	23743680
0.0	12.1641	25800	0.4502	23931472
0.0	12.2584	26000	0.4644	24118800
0.0	12.3527	26200	0.4797	24308976
0.0	12.4470	26400	0.4984	24493584
0.0	12.5413	26600	0.5124	24679264
0.0	12.6355	26800	0.4945	24861136
0.0	12.7298	27000	0.5374	25046496
0.0	12.8241	27200	0.5342	25230592
0.0	12.9184	27400	0.5432	25411904
0.0	13.0127	27600	0.5496	25595280
0.0	13.1070	27800	0.5563	25777696
0.0	13.2013	28000	0.6111	25963552
0.0	13.2956	28200	0.5734	26150464
0.0	13.3899	28400	0.5818	26335552
0.0	13.4842	28600	0.5867	26524096
0.0	13.5785	28800	0.5986	26713392
0.0	13.6728	29000	0.5962	26900464
0.0	13.7671	29200	0.6013	27087040
0.0	13.8614	29400	0.5731	27270960
0.0	13.9557	29600	0.5115	27457936
0.0	14.0500	29800	0.5246	27639216
0.0	14.1443	30000	0.5337	27829056
0.0	14.2386	30200	0.5366	28019840
0.0	14.3329	30400	0.5459	28205616
0.0	14.4272	30600	0.5455	28390464
0.0	14.5215	30800	0.5549	28571424
0.0	14.6157	31000	0.5596	28758128
0.0	14.7100	31200	0.5259	28942096
0.0	14.8043	31400	0.5383	29127440
0.0	14.8986	31600	0.5417	29310016
0.0	14.9929	31800	0.5463	29497520
0.0	15.0872	32000	0.5514	29680160
0.0	15.1815	32200	0.5603	29872080
0.0	15.2758	32400	0.5662	30060048
0.0	15.3701	32600	0.5684	30243024
0.0	15.4644	32800	0.5755	30433968
0.0	15.5587	33000	0.5767	30617936
0.0	15.6530	33200	0.5776	30802960
0.0	15.7473	33400	0.5784	30985296
0.0	15.8416	33600	0.5798	31168496
0.0	15.9359	33800	0.5820	31350688
0.0	16.0302	34000	0.5896	31530704
0.0	16.1245	34200	0.5923	31718960
0.0	16.2188	34400	0.5932	31901696
0.0	16.3131	34600	0.5964	32092528
0.0	16.4074	34800	0.6018	32279920
0.0	16.5017	35000	0.6009	32461952
0.0	16.5959	35200	0.6038	32647696
0.0	16.6902	35400	0.6065	32828656
0.0	16.7845	35600	0.6074	33016320
0.0	16.8788	35800	0.6085	33202224
0.0	16.9731	36000	0.6121	33385424
0.0	17.0674	36200	0.6134	33572672
0.0	17.1617	36400	0.6146	33759120
0.0	17.2560	36600	0.6149	33946224
0.0	17.3503	36800	0.6178	34137504
0.0	17.4446	37000	0.6199	34322448
0.0	17.5389	37200	0.6207	34506880
0.0	17.6332	37400	0.6216	34692032
0.0	17.7275	37600	0.6240	34873984
0.0	17.8218	37800	0.6267	35058576
0.0	17.9161	38000	0.6294	35245152
0.0	18.0104	38200	0.6258	35431232
0.0	18.1047	38400	0.6282	35615248
0.0	18.1990	38600	0.6322	35798688
0.0	18.2933	38800	0.6329	35984224
0.0	18.3876	39000	0.6326	36168064
0.0	18.4818	39200	0.6297	36351216
0.0	18.5761	39400	0.6315	36537456
0.0	18.6704	39600	0.6340	36723376
0.0	18.7647	39800	0.6301	36910256
0.0	18.8590	40000	0.6324	37097424

Framework versions

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