train_record_1745950249

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

• Loss: 15.1265
• Num Input Tokens Seen: 55002224

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
12.9802	0.0064	200	15.4294	277264
15.5653	0.0128	400	15.3182	548976
14.8651	0.0192	600	15.2347	826016
15.0577	0.0256	800	15.2450	1099968
14.3819	0.0320	1000	15.2481	1374672
14.8614	0.0384	1200	15.2286	1647936
14.6905	0.0448	1400	15.2245	1921648
14.7451	0.0512	1600	15.2363	2194448
12.4446	0.0576	1800	15.2256	2472048
14.8964	0.0640	2000	15.2178	2746752
16.8131	0.0704	2200	15.1959	3020144
13.821	0.0768	2400	15.2336	3296624
14.8777	0.0832	2600	15.2111	3571808
14.2003	0.0896	2800	15.2043	3847184
14.7555	0.0960	3000	15.1955	4121024
12.9845	0.1024	3200	15.1956	4396880
15.1686	0.1088	3400	15.2007	4671152
14.952	0.1152	3600	15.1925	4950800
15.3092	0.1216	3800	15.1615	5228512
14.5589	0.1280	4000	15.1821	5504608
16.1102	0.1344	4200	15.1833	5778176
16.944	0.1408	4400	15.1632	6055712
16.6215	0.1472	4600	15.1717	6331680
13.1747	0.1536	4800	15.1675	6604544
14.1834	0.1600	5000	15.1918	6882256
14.1643	0.1664	5200	15.1407	7159072
16.262	0.1728	5400	15.1633	7433136
15.0187	0.1792	5600	15.1981	7707776
15.4275	0.1856	5800	15.2009	7985472
14.5093	0.1920	6000	15.1784	8259552
15.5713	0.1985	6200	15.1697	8535952
15.0231	0.2049	6400	15.1983	8809968
13.2381	0.2113	6600	15.1305	9084016
15.0543	0.2177	6800	15.1265	9357456
14.2839	0.2241	7000	15.1470	9630608
14.1731	0.2305	7200	15.1426	9907888
14.4535	0.2369	7400	15.1830	10182048
16.5245	0.2433	7600	15.1353	10458544
14.0598	0.2497	7800	15.1353	10736144
15.952	0.2561	8000	15.1630	11010512
13.0096	0.2625	8200	15.1618	11284128
15.3546	0.2689	8400	15.1654	11556816
16.3646	0.2753	8600	15.1641	11828816
14.7599	0.2817	8800	15.1585	12104176
13.1527	0.2881	9000	15.1686	12378784
14.0753	0.2945	9200	15.1715	12654368
15.9636	0.3009	9400	15.1761	12927088
15.208	0.3073	9600	15.1809	13199552
14.3356	0.3137	9800	15.1842	13473952
15.6574	0.3201	10000	15.1568	13750288
14.857	0.3265	10200	15.1562	14025248
15.7608	0.3329	10400	15.1558	14300160
16.2262	0.3393	10600	15.1596	14577760
15.392	0.3457	10800	15.1578	14851280
13.7857	0.3521	11000	15.1574	15125104
16.3688	0.3585	11200	15.1550	15398624
15.6465	0.3649	11400	15.1580	15672384
16.6331	0.3713	11600	15.1562	15946384
14.5599	0.3777	11800	15.1567	16220112
14.0347	0.3841	12000	15.1537	16493920
15.4296	0.3905	12200	15.1600	16771376
16.117	0.3969	12400	15.1438	17046656
15.5492	0.4033	12600	15.1435	17318272
15.4808	0.4097	12800	15.1435	17591696
15.5101	0.4161	13000	15.1441	17864256
14.4989	0.4225	13200	15.1461	18137984
13.6671	0.4289	13400	15.1443	18413504
14.6588	0.4353	13600	15.1439	18690528
17.4437	0.4417	13800	15.1473	18966352
14.0331	0.4481	14000	15.1442	19242160
13.0993	0.4545	14200	15.1441	19518832
13.6133	0.4609	14400	15.1444	19795920
15.4418	0.4673	14600	15.1433	20073168
13.4136	0.4737	14800	15.1449	20349056
14.7192	0.4801	15000	15.1442	20622896
15.6836	0.4865	15200	15.1551	20896768
13.27	0.4929	15400	15.1552	21171376
14.9471	0.4993	15600	15.1545	21447568
16.4186	0.5057	15800	15.1562	21722256
14.1038	0.5121	16000	15.1550	21998320
15.4738	0.5185	16200	15.1552	22273616
13.9429	0.5249	16400	15.1555	22549280
13.12	0.5313	16600	15.1553	22823984
14.7638	0.5377	16800	15.1552	23098384
15.258	0.5441	17000	15.1557	23371136
14.6926	0.5505	17200	15.1556	23647856
15.8429	0.5569	17400	15.1555	23921008
15.6753	0.5633	17600	15.1550	24194480
16.3745	0.5697	17800	15.1500	24469312
16.2398	0.5761	18000	15.1560	24743360
15.6079	0.5825	18200	15.1588	25020352
15.0036	0.5890	18400	15.1589	25295920
15.0048	0.5954	18600	15.1587	25571232
15.138	0.6018	18800	15.1586	25847664
15.5366	0.6082	19000	15.1587	26125328
13.933	0.6146	19200	15.1587	26404064
14.128	0.6210	19400	15.1587	26677504
15.8011	0.6274	19600	15.1587	26952544
15.3115	0.6338	19800	15.1588	27226896
14.7545	0.6402	20000	15.1588	27501216
14.6687	0.6466	20200	15.1588	27776624
15.6931	0.6530	20400	15.1588	28051872
13.5465	0.6594	20600	15.1588	28325632
13.7558	0.6658	20800	15.1588	28598784
15.2542	0.6722	21000	15.1588	28874800
16.6519	0.6786	21200	15.1597	29151312
17.4344	0.6850	21400	15.1596	29425936
12.8174	0.6914	21600	15.1603	29702784
15.1029	0.6978	21800	15.1595	29979824
14.3423	0.7042	22000	15.1599	30256128
15.6432	0.7106	22200	15.1586	30528032
14.3856	0.7170	22400	15.1591	30803904
15.1348	0.7234	22600	15.1591	31077632
15.322	0.7298	22800	15.1594	31354544
14.2641	0.7362	23000	15.1594	31626736
16.7093	0.7426	23200	15.1601	31901472
16.5769	0.7490	23400	15.1601	32179968
15.7483	0.7554	23600	15.1596	32457728
15.3673	0.7618	23800	15.1589	32732288
13.9073	0.7682	24000	15.1598	33007504
14.3764	0.7746	24200	15.1597	33281968
14.7825	0.7810	24400	15.1586	33558736
14.8331	0.7874	24600	15.1586	33830832
14.3912	0.7938	24800	15.1594	34104944
14.5176	0.8002	25000	15.1601	34381536
15.3626	0.8066	25200	15.1596	34654672
15.0243	0.8130	25400	15.1601	34931520
15.9599	0.8194	25600	15.1601	35206448
14.9095	0.8258	25800	15.1596	35482800
16.1984	0.8322	26000	15.1599	35756816
15.5676	0.8386	26200	15.1601	36031296
15.3677	0.8450	26400	15.1601	36307968
16.4824	0.8514	26600	15.1601	36580432
14.258	0.8578	26800	15.1596	36855328
15.7652	0.8642	27000	15.1601	37133072
14.3243	0.8706	27200	15.1601	37404464
15.1033	0.8770	27400	15.1601	37675456
14.5413	0.8834	27600	15.1601	37951616
13.5684	0.8898	27800	15.1593	38225840
15.5201	0.8962	28000	15.1593	38498736
15.159	0.9026	28200	15.1593	38771760
16.9553	0.9090	28400	15.1597	39045824
12.5237	0.9154	28600	15.1597	39320736
14.7548	0.9218	28800	15.1597	39594816
14.4744	0.9282	29000	15.1597	39870432
16.6983	0.9346	29200	15.1597	40144672
13.4943	0.9410	29400	15.1597	40420752
15.7357	0.9474	29600	15.1597	40696672
13.7294	0.9538	29800	15.1597	40970096
14.1421	0.9602	30000	15.1597	41245904
15.3599	0.9666	30200	15.1597	41519232
14.5299	0.9730	30400	15.1597	41791520
15.7709	0.9795	30600	15.1597	42066928
14.3146	0.9859	30800	15.1597	42339616
15.0257	0.9923	31000	15.1597	42616352
15.7705	0.9987	31200	15.1597	42892688
15.5756	1.0051	31400	15.1597	43167792
15.867	1.0115	31600	15.1597	43444592
13.7023	1.0179	31800	15.1597	43719328
13.9696	1.0243	32000	15.1597	43994064
14.3737	1.0307	32200	15.1597	44269712
14.9996	1.0371	32400	15.1597	44545408
12.366	1.0435	32600	15.1597	44819808
14.4042	1.0499	32800	15.1597	45097904
15.2717	1.0563	33000	15.1597	45376272
14.2669	1.0627	33200	15.1597	45647824
14.3355	1.0691	33400	15.1597	45922032
14.4671	1.0755	33600	15.1597	46197840
13.329	1.0819	33800	15.1597	46474848
15.1515	1.0883	34000	15.1597	46749824
13.4913	1.0947	34200	15.1597	47023856
15.6598	1.1011	34400	15.1597	47301520
14.2071	1.1075	34600	15.1597	47574864
16.7937	1.1139	34800	15.1597	47853888
14.6934	1.1203	35000	15.1597	48129792
15.649	1.1267	35200	15.1597	48405024
13.6601	1.1331	35400	15.1597	48678592
14.6172	1.1395	35600	15.1597	48954048
15.9964	1.1459	35800	15.1597	49232480
15.4332	1.1523	36000	15.1597	49505040
14.3303	1.1587	36200	15.1597	49778864
15.196	1.1651	36400	15.1597	50051632
14.1885	1.1715	36600	15.1597	50325888
13.9557	1.1779	36800	15.1597	50601136
14.7712	1.1843	37000	15.1597	50876992
14.3635	1.1907	37200	15.1597	51153296
16.9773	1.1971	37400	15.1597	51427552
17.6143	1.2035	37600	15.1597	51707088
13.8467	1.2099	37800	15.1597	51981712
14.8764	1.2163	38000	15.1597	52254352
16.6284	1.2227	38200	15.1597	52529584
14.2186	1.2291	38400	15.1597	52803776
15.0678	1.2355	38600	15.1597	53078736
14.9189	1.2419	38800	15.1597	53352672
13.976	1.2483	39000	15.1597	53628768
13.8707	1.2547	39200	15.1597	53905216
13.4688	1.2611	39400	15.1597	54178832
13.2421	1.2675	39600	15.1597	54454880
14.5777	1.2739	39800	15.1597	54727600
16.7093	1.2803	40000	15.1597	55002224

Framework versions

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