train_boolq_1745950282

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: 1.5235
• 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
0.9126	0.0943	200	1.6075	186768
1.8665	0.1886	400	1.5869	369808
2.3282	0.2829	600	1.5783	554928
1.2111	0.3772	800	1.5787	746560
0.6563	0.4715	1000	1.5645	932848
1.6293	0.5658	1200	1.5633	1116128
1.9335	0.6601	1400	1.5533	1299664
2.187	0.7544	1600	1.5488	1481856
1.1897	0.8487	1800	1.5607	1672160
1.521	0.9430	2000	1.5519	1860608
1.5792	1.0372	2200	1.5547	2047984
1.9829	1.1315	2400	1.5497	2230960
2.2104	1.2258	2600	1.5496	2417664
1.5622	1.3201	2800	1.5517	2600368
0.5567	1.4144	3000	1.5504	2786848
1.7935	1.5087	3200	1.5425	2972672
0.828	1.6030	3400	1.5452	3154640
1.7858	1.6973	3600	1.5419	3339328
1.5979	1.7916	3800	1.5441	3522384
1.5046	1.8859	4000	1.5455	3712352
0.8123	1.9802	4200	1.5366	3899328
1.9172	2.0745	4400	1.5408	4085888
1.7031	2.1688	4600	1.5413	4271936
1.2118	2.2631	4800	1.5362	4456320
1.3216	2.3574	5000	1.5418	4638512
2.1133	2.4517	5200	1.5425	4830688
1.5954	2.5460	5400	1.5369	5016480
1.5956	2.6403	5600	1.5407	5204048
1.2136	2.7346	5800	1.5380	5383984
2.0436	2.8289	6000	1.5334	5574016
2.1082	2.9231	6200	1.5342	5761616
1.6868	3.0174	6400	1.5428	5948128
1.6432	3.1117	6600	1.5433	6134304
1.2227	3.2060	6800	1.5346	6319616
2.4769	3.3003	7000	1.5341	6505744
0.7089	3.3946	7200	1.5427	6692208
0.931	3.4889	7400	1.5286	6875616
1.6422	3.5832	7600	1.5338	7059472
2.4477	3.6775	7800	1.5321	7243472
2.3196	3.7718	8000	1.5354	7428048
0.9875	3.8661	8200	1.5305	7611184
0.8705	3.9604	8400	1.5299	7796112
1.726	4.0547	8600	1.5377	7979520
1.05	4.1490	8800	1.5366	8167776
0.8927	4.2433	9000	1.5298	8355856
1.4016	4.3376	9200	1.5397	8543120
1.2307	4.4319	9400	1.5363	8727088
1.3127	4.5262	9600	1.5346	8914992
1.3066	4.6205	9800	1.5394	9095040
1.7302	4.7148	10000	1.5311	9283072
1.1526	4.8091	10200	1.5369	9467600
1.8063	4.9033	10400	1.5314	9653456
2.1182	4.9976	10600	1.5371	9841232
0.6087	5.0919	10800	1.5387	10025504
0.741	5.1862	11000	1.5376	10216464
3.1286	5.2805	11200	1.5334	10402448
1.0825	5.3748	11400	1.5321	10586976
0.864	5.4691	11600	1.5356	10770896
1.7066	5.5634	11800	1.5378	10959424
1.2288	5.6577	12000	1.5421	11146816
1.8905	5.7520	12200	1.5380	11328528
1.1623	5.8463	12400	1.5419	11515600
1.6314	5.9406	12600	1.5384	11697056
0.4464	6.0349	12800	1.5378	11884336
1.1535	6.1292	13000	1.5300	12074128
0.2332	6.2235	13200	1.5386	12258064
1.6191	6.3178	13400	1.5347	12443248
0.9751	6.4121	13600	1.5278	12626480
1.8843	6.5064	13800	1.5368	12813808
1.5336	6.6007	14000	1.5303	12998256
0.8781	6.6950	14200	1.5382	13180928
1.9583	6.7893	14400	1.5305	13364368
2.0035	6.8835	14600	1.5235	13552272
1.9377	6.9778	14800	1.5285	13735904
2.7577	7.0721	15000	1.5361	13924000
2.4858	7.1664	15200	1.5375	14113184
1.8379	7.2607	15400	1.5328	14295568
0.5395	7.3550	15600	1.5352	14480560
1.4666	7.4493	15800	1.5322	14664736
2.6556	7.5436	16000	1.5277	14852128
0.8046	7.6379	16200	1.5363	15033840
1.0626	7.7322	16400	1.5361	15219136
1.4575	7.8265	16600	1.5316	15404160
2.1096	7.9208	16800	1.5287	15589632
1.2183	8.0151	17000	1.5347	15781760
0.7056	8.1094	17200	1.5361	15967648
1.0756	8.2037	17400	1.5345	16155248
1.6033	8.2980	17600	1.5316	16343648
1.2349	8.3923	17800	1.5410	16523360
1.3093	8.4866	18000	1.5324	16709008
1.3615	8.5809	18200	1.5336	16893648
0.632	8.6752	18400	1.5273	17079824
1.8603	8.7694	18600	1.5302	17265072
2.1083	8.8637	18800	1.5434	17445904
1.4074	8.9580	19000	1.5448	17631504
0.9535	9.0523	19200	1.5377	17818512
0.6912	9.1466	19400	1.5420	18005200
1.7746	9.2409	19600	1.5344	18190416
2.0576	9.3352	19800	1.5400	18373200
2.5187	9.4295	20000	1.5314	18556672
2.6635	9.5238	20200	1.5391	18742816
2.1528	9.6181	20400	1.5346	18930224
1.7716	9.7124	20600	1.5400	19115456
1.355	9.8067	20800	1.5383	19296016
0.5675	9.9010	21000	1.5383	19482416
0.974	9.9953	21200	1.5275	19668640
1.1068	10.0896	21400	1.5334	19860880
1.0851	10.1839	21600	1.5333	20052672
0.9046	10.2782	21800	1.5318	20236224
1.5511	10.3725	22000	1.5293	20421632
2.4675	10.4668	22200	1.5286	20608320
1.1624	10.5611	22400	1.5350	20788112
1.0901	10.6554	22600	1.5330	20969744
0.9	10.7496	22800	1.5276	21151648
2.3174	10.8439	23000	1.5259	21335600
1.3619	10.9382	23200	1.5372	21522352
2.7212	11.0325	23400	1.5358	21709568
1.3666	11.1268	23600	1.5326	21894592
1.3113	11.2211	23800	1.5271	22079344
1.0245	11.3154	24000	1.5336	22269152
2.2966	11.4097	24200	1.5339	22451760
1.6225	11.5040	24400	1.5459	22639312
0.5084	11.5983	24600	1.5342	22821728
0.9625	11.6926	24800	1.5335	23005696
1.3978	11.7869	25000	1.5367	23192112
1.7703	11.8812	25200	1.5308	23373840
1.3747	11.9755	25400	1.5273	23559968
1.1913	12.0698	25600	1.5376	23743680
2.9794	12.1641	25800	1.5405	23931472
0.6398	12.2584	26000	1.5336	24118800
1.9674	12.3527	26200	1.5306	24308976
0.641	12.4470	26400	1.5383	24493584
0.9799	12.5413	26600	1.5285	24679264
1.8837	12.6355	26800	1.5345	24861136
1.7178	12.7298	27000	1.5277	25046496
2.0154	12.8241	27200	1.5377	25230592
2.2121	12.9184	27400	1.5383	25411904
1.9462	13.0127	27600	1.5341	25595280
1.892	13.1070	27800	1.5425	25777696
0.7727	13.2013	28000	1.5292	25963552
0.6919	13.2956	28200	1.5313	26150464
2.4513	13.3899	28400	1.5388	26335552
1.9732	13.4842	28600	1.5330	26524096
2.0418	13.5785	28800	1.5242	26713392
0.6923	13.6728	29000	1.5337	26900464
1.8188	13.7671	29200	1.5407	27087040
1.383	13.8614	29400	1.5411	27270960
2.4062	13.9557	29600	1.5326	27457936
2.2125	14.0500	29800	1.5275	27639216
1.6523	14.1443	30000	1.5383	27829056
1.2126	14.2386	30200	1.5311	28019840
0.7117	14.3329	30400	1.5315	28205616
2.046	14.4272	30600	1.5307	28390464
0.8394	14.5215	30800	1.5439	28571424
0.5962	14.6157	31000	1.5324	28758128
1.3903	14.7100	31200	1.5316	28942096
1.9185	14.8043	31400	1.5297	29127440
0.3431	14.8986	31600	1.5316	29310016
1.5252	14.9929	31800	1.5402	29497520
2.2414	15.0872	32000	1.5419	29680160
1.1793	15.1815	32200	1.5367	29872080
2.1016	15.2758	32400	1.5358	30060048
2.7248	15.3701	32600	1.5426	30243024
1.5292	15.4644	32800	1.5275	30433968
0.88	15.5587	33000	1.5373	30617936
0.3452	15.6530	33200	1.5300	30802960
0.7676	15.7473	33400	1.5313	30985296
0.8678	15.8416	33600	1.5355	31168496
1.3862	15.9359	33800	1.5326	31350688
1.3233	16.0302	34000	1.5336	31530704
1.5892	16.1245	34200	1.5307	31718960
0.6211	16.2188	34400	1.5267	31901696
1.4859	16.3131	34600	1.5274	32092528
2.7674	16.4074	34800	1.5282	32279920
2.014	16.5017	35000	1.5344	32461952
0.8514	16.5959	35200	1.5360	32647696
1.6113	16.6902	35400	1.5324	32828656
0.26	16.7845	35600	1.5334	33016320
1.6374	16.8788	35800	1.5295	33202224
1.3047	16.9731	36000	1.5368	33385424
2.1828	17.0674	36200	1.5384	33572672
1.343	17.1617	36400	1.5347	33759120
1.7243	17.2560	36600	1.5392	33946224
1.2776	17.3503	36800	1.5344	34137504
2.3692	17.4446	37000	1.5323	34322448
1.059	17.5389	37200	1.5340	34506880
2.9969	17.6332	37400	1.5392	34692032
1.4363	17.7275	37600	1.5348	34873984
2.3506	17.8218	37800	1.5300	35058576
0.3766	17.9161	38000	1.5268	35245152
1.7634	18.0104	38200	1.5272	35431232
1.4477	18.1047	38400	1.5272	35615248
1.3816	18.1990	38600	1.5272	35798688
2.7273	18.2933	38800	1.5272	35984224
1.4794	18.3876	39000	1.5272	36168064
0.3987	18.4818	39200	1.5272	36351216
0.853	18.5761	39400	1.5272	36537456
0.8147	18.6704	39600	1.5272	36723376
0.4624	18.7647	39800	1.5272	36910256
0.7265	18.8590	40000	1.5272	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