learning_rate: 0.001
train_batch_size: 32
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 512
optimizer: Use OptimizerNames.ADAMW_APEX_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: warmup_stable_decay
lr_scheduler_warmup_steps: 500
training_steps: 140000

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
3.8584	0.0048	500	3.6764	262104320
3.1457	0.0095	1000	3.0411	524218944
2.9472	0.0143	1500	2.8491	786331040
2.8357	0.0191	2000	2.7469	1048424032
2.7564	0.0239	2500	2.6824	1310500000
2.719	0.0286	3000	2.6300	1572599776
2.664	0.0334	3500	2.5908	1834732864
2.6351	0.0382	4000	2.5541	2096832320
2.5897	0.0429	4500	2.5187	2358960576
2.5672	0.0477	5000	2.4908	2621089856
2.5412	0.0525	5500	2.4639	2883166048
2.5194	0.0572	6000	2.4423	3145255744
2.5105	0.0620	6500	2.4206	3407356352
2.4816	0.0668	7000	2.4051	3669468320
2.4688	0.0716	7500	2.3871	3931543616
2.4446	0.0763	8000	2.3720	4193649312
2.4318	0.0811	8500	2.3614	4455747616
2.4292	0.0859	9000	2.3485	4717856864
2.4151	0.0906	9500	2.3371	4979955776
2.4056	0.0954	10000	2.3263	5242058496
2.3912	0.1002	10500	2.3170	5504131840
2.3781	0.1049	11000	2.3075	5766222432
2.3776	0.1097	11500	2.2976	6028313408
2.3759	0.1145	12000	2.2900	6290430912
2.3501	0.1193	12500	2.2824	6552526688
2.3572	0.1240	13000	2.2758	6814626336
2.3451	0.1288	13500	2.2677	7076718080
2.3437	0.1336	14000	2.2614	7338830528
2.328	0.1383	14500	2.2562	7600938432
2.3288	0.1431	15000	2.2487	7863022432
2.3259	0.1479	15500	2.2430	8125131456
2.3103	0.1526	16000	2.2366	8387218560
2.3137	0.1574	16500	2.2321	8649321536
2.3148	0.1622	17000	2.2286	8911431360
2.3076	0.1670	17500	2.2226	9173533056
2.2963	0.1717	18000	2.2166	9435637536
2.3052	0.1765	18500	2.2123	9697738752
2.2941	0.1813	19000	2.2090	9959851776
2.2888	0.1860	19500	2.2118	10221963136
2.2894	0.1908	20000	2.2017	10484059168
2.2882	0.1956	20500	2.1974	10746164768
2.2677	0.2003	21000	2.1927	11008255872
2.2577	0.2051	21500	2.1901	11270362240
2.2725	0.2099	22000	2.1856	11532457408
2.2519	0.2147	22500	2.1839	11794558656
2.266	0.2194	23000	2.1793	12056655104
2.2531	0.2242	23500	2.1767	12318747360
2.2522	0.2290	24000	2.1732	12580853504
2.2604	0.2337	24500	2.1710	12842964128
2.253	0.2385	25000	2.1673	13105069824
2.2388	0.2433	25500	2.1654	13367175456
2.2511	0.2480	26000	2.1629	13629260960
2.2453	0.2528	26500	2.1592	13891381408
2.2302	0.2576	27000	2.1568	14153506688
2.2305	0.2624	27500	2.1540	14415584288
2.2285	0.2671	28000	2.1511	14677696896
2.23	0.2719	28500	2.1498	14939780320
2.2136	0.2767	29000	2.1479	15201894176
2.2333	0.2814	29500	2.1446	15463988928
2.2241	0.2862	30000	2.1426	15726072896
2.2318	0.2910	30500	2.1404	15988180544
2.2156	0.2957	31000	2.1403	16250292672
2.2184	0.3005	31500	2.1374	16512390176
2.2261	0.3053	32000	2.1346	16774478112
2.2091	0.3101	32500	2.1335	17036582752
2.2222	0.3148	33000	2.1309	17298706464
2.2172	0.3196	33500	2.1288	17560813216
2.2003	0.3244	34000	2.1275	17822921152
2.2152	0.3291	34500	2.1281	18085018528
2.2076	0.3339	35000	2.1236	18347147168
2.2005	0.3387	35500	2.1224	18609222656
2.1975	0.3434	36000	2.1199	18871341760
2.201	0.3482	36500	2.1179	19133442016
2.1968	0.3530	37000	2.1169	19395541920
2.204	0.3578	37500	2.1144	19657633088
2.1979	0.3625	38000	2.1132	19919735840
2.1898	0.3673	38500	2.1110	20181833600
2.1859	0.3721	39000	2.1110	20443920960
2.188	0.3768	39500	2.1096	20706018592
2.1932	0.3816	40000	2.1082	20968112960
2.1933	0.3864	40500	2.1045	21230223584
2.1907	0.3911	41000	2.1057	21492310496
2.1806	0.3959	41500	2.1030	21754409120
2.1834	0.4007	42000	2.1014	22016505376
2.1914	0.4055	42500	2.1016	22278605888
2.1932	0.4102	43000	2.0990	22540715296
2.2209	0.4150	43500	2.1086	22802815776
2.1856	0.4198	44000	2.0981	23064909408
2.1823	0.4245	44500	2.0960	23327017760
2.1862	0.4293	45000	2.0936	23589115072
2.182	0.4341	45500	2.0927	23851210336
2.1729	0.4388	46000	2.0917	24113281184
2.177	0.4436	46500	2.0904	24375397792
2.1674	0.4484	47000	2.0890	24637513248
2.1608	0.4532	47500	2.0887	24899608000
2.1808	0.4579	48000	2.0938	25161718656
2.1811	0.4627	48500	2.0870	25423801984
2.1621	0.4675	49000	2.0852	25685912544
2.1722	0.4722	49500	2.0832	25948022560
2.1745	0.4770	50000	2.0824	26210133120
2.1529	0.4818	50500	2.0812	26472227840
2.169	0.4865	51000	2.0815	26734340064
2.1738	0.4913	51500	2.0796	26996432960
2.169	0.4961	52000	2.0802	27258524544
2.1557	0.5009	52500	2.0776	27520636736
2.1765	0.5056	53000	2.0828	27782732608
2.1616	0.5104	53500	2.0767	28044839456
2.1569	0.5152	54000	2.0758	28306946368
2.1561	0.5199	54500	2.0746	28569047936
2.1554	0.5247	55000	2.0725	28831152896
2.1505	0.5295	55500	2.0716	29093257888
2.1491	0.5342	56000	2.0714	29355372320
2.1471	0.5390	56500	2.0707	29617485024
2.1465	0.5438	57000	2.0692	29879599072
2.1511	0.5486	57500	2.0681	30141698752
2.1456	0.5533	58000	2.0688	30403788864
2.1591	0.5581	58500	2.0664	30665890560
2.1508	0.5629	59000	2.0671	30927998464
2.1466	0.5676	59500	2.0664	31190116608
2.1457	0.5724	60000	2.0640	31452217632
2.1496	0.5772	60500	2.0636	31714314848
2.1418	0.5819	61000	2.0649	31976431072
2.1477	0.5867	61500	2.0638	32238532768
2.137	0.5915	62000	2.0610	32500617568
2.1415	0.5963	62500	2.0606	32762704928
2.1459	0.6010	63000	2.0603	33024820736
2.1389	0.6058	63500	2.0586	33286935872
2.1367	0.6106	64000	2.0588	33549034848
2.147	0.6153	64500	2.0593	33811149696
2.1415	0.6201	65000	2.0580	34073267168
2.1426	0.6249	65500	2.0569	34335361632
2.1483	0.6296	66000	2.0545	34597457472
2.1409	0.6344	66500	2.0548	34859578368
2.1368	0.6392	67000	2.0555	35121682240
2.1366	0.6440	67500	2.0543	35383796224
2.137	0.6487	68000	2.0545	35645894016
2.1342	0.6535	68500	2.0521	35907993664
2.1388	0.6583	69000	2.0507	36170105088
2.1339	0.6630	69500	2.0502	36432197248
2.118	0.6678	70000	2.0507	36694303392
2.134	0.6726	70500	2.0488	36956418880
2.1461	0.6773	71000	2.0492	37218518048
2.1361	0.6821	71500	2.0479	37480608672
2.1369	0.6869	72000	2.0480	37742700480
2.1369	0.6917	72500	2.0477	38004789088
2.1379	0.6964	73000	2.0459	38266876896
2.1235	0.7012	73500	2.0458	38528990144
2.127	0.7060	74000	2.0454	38791103008
2.1362	0.7107	74500	2.0453	39053184384
2.1096	0.7155	75000	2.0448	39315259072
2.1273	0.7203	75500	2.0433	39577349376
2.1325	0.7250	76000	2.0433	39839459232
2.1261	0.7298	76500	2.0422	40101539296
2.1284	0.7346	77000	2.0418	40363651328
2.1152	0.7394	77500	2.0417	40625741536
2.1353	0.7441	78000	2.0413	40887844000
2.1214	0.7489	78500	2.0390	41149941248
2.1184	0.7537	79000	2.0395	41412054816
2.1199	0.7584	79500	2.0389	41674139648
2.129	0.7632	80000	2.0402	41936220160
2.1247	0.7680	80500	2.0385	42198339200
2.1147	0.7727	81000	2.0367	42460451136
2.1265	0.7775	81500	2.0365	42722562592
2.1254	0.7823	82000	2.0358	42984652928
2.1121	0.7871	82500	2.0354	43246727552
2.1235	0.7918	83000	2.0352	43508842912
2.1192	0.7966	83500	2.0352	43770947360
2.1303	0.8014	84000	2.0382	44033053408
2.1247	0.8061	84500	2.0337	44295182016
2.125	0.8109	85000	2.0327	44557289600
2.1218	0.8157	85500	2.0317	44819414752
2.1122	0.8204	86000	2.0312	45081505024
2.1293	0.8252	86500	2.0313	45343592896
2.1161	0.8300	87000	2.0335	45605708608
2.1237	0.8348	87500	2.0365	45867808448
2.1206	0.8395	88000	2.0313	46129884928
2.1191	0.8443	88500	2.0305	46391967104
2.1075	0.8491	89000	2.0287	46654069856
2.1109	0.8538	89500	2.0292	46916134592
2.1208	0.8586	90000	2.0290	47178248768
2.1056	0.8634	90500	2.0283	47440355904
2.1071	0.8681	91000	2.0275	47702460032
2.1069	0.8729	91500	2.0270	47964538176
2.1003	0.8777	92000	2.0305	48226649920
2.102	0.8825	92500	2.0255	48488770560
2.1141	0.8872	93000	2.0259	48750829152
2.1241	0.8920	93500	2.0255	49012940480
2.1154	0.8968	94000	2.0230	49275046560
2.0967	0.9015	94500	2.0232	49537142912
2.1101	0.9063	95000	2.0240	49799231680
2.1087	0.9111	95500	2.0230	50061331936
2.1011	0.9158	96000	2.0229	50323443136
2.1052	0.9206	96500	2.0226	50585537568
2.1024	0.9254	97000	2.0225	50847631776
2.1032	0.9302	97500	2.0224	51109732512
2.0984	0.9349	98000	2.0211	51371830432
2.1133	0.9397	98500	2.0206	51633931744
2.1038	0.9445	99000	2.0205	51896030848
2.1021	0.9492	99500	2.0222	52158147168
2.1116	0.9540	100000	2.0211	52420247968
2.1026	0.9588	100500	2.0197	52682343360
2.1031	0.9635	101000	2.0206	52944447232
2.1127	0.9683	101500	2.0185	53206553568
2.0992	0.9731	102000	2.0180	53468658336
2.0968	0.9779	102500	2.0211	53730772800
2.092	0.9826	103000	2.0180	53992863264
2.1016	0.9874	103500	2.0163	54254966464
2.1098	0.9922	104000	2.0174	54517083360
2.1103	0.9969	104500	2.0176	54779191328
2.0879	1.0017	105000	2.0179	55041423104
2.0964	1.0065	105500	2.0158	55303527552
2.0989	1.0112	106000	2.0157	55565623584
2.1079	1.0160	106500	2.0186	55827708736
2.1069	1.0208	107000	2.0148	56089794304
2.0987	1.0256	107500	2.0147	56351910752
2.095	1.0303	108000	2.0153	56614015392
2.097	1.0351	108500	2.0137	56876128320
2.099	1.0399	109000	2.0129	57138237248
2.0952	1.0446	109500	2.0124	57400333280
2.0895	1.0494	110000	2.0132	57662439520
2.0945	1.0542	110500	2.0113	57924525696
2.0978	1.0589	111000	2.0107	58186625280
2.0873	1.0637	111500	2.0106	58448739744
2.0768	1.0685	112000	2.0110	58710836096
2.0901	1.0733	112500	2.0121	58972918688
2.0889	1.0780	113000	2.0116	59235047232
2.0985	1.0828	113500	2.0100	59497148000
2.1012	1.0876	114000	2.0105	59759249088
2.0837	1.0923	114500	2.0083	60021363392
2.0992	1.0971	115000	2.0077	60283464768
2.0916	1.1019	115500	2.0105	60545534656
2.1125	1.1066	116000	2.0121	60807636160
2.0889	1.1114	116500	2.0075	61069752768
2.0879	1.1162	117000	2.0054	61331831488
2.0759	1.1210	117500	2.0039	61593927520
2.0801	1.1257	118000	2.0020	61856020192
2.0836	1.1305	118500	2.0045	62118128864
2.0858	1.1353	119000	2.0006	62380238112
2.0836	1.1400	119500	1.9990	62642339520
2.0956	1.1448	120000	2.0033	62904447680
2.0796	1.1496	120500	1.9961	63166554368
2.077	1.1543	121000	1.9940	63428647904
2.0938	1.1591	121500	1.9930	63690757024
2.072	1.1639	122000	1.9911	63952872768
2.0684	1.1687	122500	1.9892	64214942816
2.0564	1.1734	123000	1.9881	64477051392
2.066	1.1782	123500	1.9865	64739145440
2.0639	1.1830	124000	1.9843	65001257824
2.0675	1.1877	124500	1.9832	65263378112
2.0692	1.1925	125000	1.9822	65525493280
2.0645	1.1973	125500	1.9799	65787582144
2.0556	1.2020	126000	1.9784	66049692768
2.0581	1.2068	126500	1.9767	66311770112
2.0633	1.2116	127000	1.9741	66573856704
2.0625	1.2164	127500	1.9729	66835970592
2.0557	1.2211	128000	1.9715	67098059328
2.0475	1.2259	128500	1.9702	67360131616
2.0558	1.2307	129000	1.9687	67622231264
2.0405	1.2354	129500	1.9668	67884327168
2.038	1.2402	130000	1.9652	68146442176
2.0338	1.2450	130500	1.9644	68408534464
2.0506	1.2497	131000	1.9626	68670633664
2.0432	1.2545	131500	1.9614	68932735776
2.0361	1.2593	132000	1.9603	69194840608
2.0347	1.2641	132500	1.9590	69456943424
2.0354	1.2688	133000	1.9577	69719047840
2.0419	1.2736	133500	1.9569	69981165632
2.0345	1.2784	134000	1.9554	70243253472
2.0393	1.2831	134500	1.9546	70505353504
2.0401	1.2879	135000	1.9541	70767457344
2.037	1.2927	135500	1.9531	71029545920
2.0302	1.2974	136000	1.9527	71291638272
2.0173	1.3022	136500	1.9525	71553726688
2.0287	1.3070	137000	1.9519	71815816608
2.0254	1.3118	137500	1.9515	72077935168
2.0286	1.3165	138000	1.9512	72340003200
2.0285	1.3213	138500	1.9511	72602127392
2.0353	1.3261	139000	1.9511	72864248896
2.0366	1.3308	139500	1.9510	73126349984
2.0315	1.3356	140000	1.9510	73388446624

Framework versions

Transformers 4.50.3
Pytorch 2.6.0+cu126
Datasets 3.5.0
Tokenizers 0.21.1

Azrail/smallm_350

smallm_350

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Azrail/
smallm_350