train_record_1745950247

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: 0.4941
• 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: 0.3
• 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
1.2371	0.0064	200	1.5827	277264
1.3444	0.0128	400	1.2721	548976
1.5009	0.0192	600	1.2182	826016
0.8678	0.0256	800	1.1834	1099968
1.1076	0.0320	1000	1.0605	1374672
0.9496	0.0384	1200	1.0762	1647936
1.0765	0.0448	1400	0.9992	1921648
0.8033	0.0512	1600	1.0182	2194448
0.9584	0.0576	1800	1.1629	2472048
1.1006	0.0640	2000	0.9536	2746752
1.0743	0.0704	2200	0.8952	3020144
1.0446	0.0768	2400	0.8950	3296624
0.9012	0.0832	2600	0.9015	3571808
1.1622	0.0896	2800	0.8872	3847184
0.9041	0.0960	3000	0.8430	4121024
0.8787	0.1024	3200	0.8599	4396880
0.8314	0.1088	3400	0.8541	4671152
0.7955	0.1152	3600	0.8506	4950800
0.7693	0.1216	3800	0.8173	5228512
0.8709	0.1280	4000	0.8573	5504608
0.9853	0.1344	4200	0.8288	5778176
1.0905	0.1408	4400	0.8487	6055712
0.846	0.1472	4600	0.8267	6331680
0.7685	0.1536	4800	0.8237	6604544
0.8457	0.1600	5000	0.8814	6882256
0.9768	0.1664	5200	0.7992	7159072
0.8985	0.1728	5400	0.7898	7433136
0.9597	0.1792	5600	0.7926	7707776
0.7812	0.1856	5800	0.7972	7985472
0.8175	0.1920	6000	0.8064	8259552
0.8294	0.1985	6200	0.7868	8535952
0.7955	0.2049	6400	0.7671	8809968
0.8651	0.2113	6600	0.7742	9084016
0.8679	0.2177	6800	0.7769	9357456
0.7892	0.2241	7000	0.7837	9630608
0.7311	0.2305	7200	0.7955	9907888
0.7695	0.2369	7400	0.7727	10182048
0.9169	0.2433	7600	0.7677	10458544
0.7497	0.2497	7800	0.7552	10736144
0.8553	0.2561	8000	0.7171	11010512
0.6293	0.2625	8200	0.6951	11284128
0.6334	0.2689	8400	0.6879	11556816
0.8279	0.2753	8600	0.7055	11828816
0.6041	0.2817	8800	0.6626	12104176
0.6143	0.2881	9000	0.6640	12378784
0.8597	0.2945	9200	0.6497	12654368
0.6403	0.3009	9400	0.6553	12927088
0.6487	0.3073	9600	0.6484	13199552
0.5817	0.3137	9800	0.6355	13473952
0.6924	0.3201	10000	0.6559	13750288
0.5349	0.3265	10200	0.6239	14025248
0.556	0.3329	10400	0.6419	14300160
0.5425	0.3393	10600	0.6385	14577760
0.6794	0.3457	10800	0.6185	14851280
0.5348	0.3521	11000	0.6385	15125104
0.6781	0.3585	11200	0.6159	15398624
0.8133	0.3649	11400	0.6244	15672384
0.5864	0.3713	11600	0.6171	15946384
0.3912	0.3777	11800	0.6202	16220112
0.6575	0.3841	12000	0.6192	16493920
0.5405	0.3905	12200	0.5922	16771376
0.7828	0.3969	12400	0.5894	17046656
0.4658	0.4033	12600	0.5989	17318272
0.7755	0.4097	12800	0.6354	17591696
0.5444	0.4161	13000	0.6029	17864256
0.5456	0.4225	13200	0.5986	18137984
0.7185	0.4289	13400	0.5926	18413504
0.6642	0.4353	13600	0.5815	18690528
0.6459	0.4417	13800	0.5842	18966352
0.5672	0.4481	14000	0.5744	19242160
0.5367	0.4545	14200	0.5751	19518832
0.5516	0.4609	14400	0.5729	19795920
0.5055	0.4673	14600	0.5889	20073168
0.399	0.4737	14800	0.5754	20349056
0.5499	0.4801	15000	0.5825	20622896
0.5207	0.4865	15200	0.5702	20896768
0.5521	0.4929	15400	0.5712	21171376
0.5371	0.4993	15600	0.5752	21447568
0.6351	0.5057	15800	0.5704	21722256
0.5437	0.5121	16000	0.5732	21998320
0.5628	0.5185	16200	0.5639	22273616
0.626	0.5249	16400	0.5587	22549280
0.5518	0.5313	16600	0.5643	22823984
0.4703	0.5377	16800	0.5626	23098384
0.5579	0.5441	17000	0.5629	23371136
0.4205	0.5505	17200	0.5602	23647856
0.5438	0.5569	17400	0.5586	23921008
0.4923	0.5633	17600	0.5529	24194480
0.4893	0.5697	17800	0.5554	24469312
0.5432	0.5761	18000	0.5534	24743360
0.5548	0.5825	18200	0.5488	25020352
0.5781	0.5890	18400	0.5530	25295920
0.5579	0.5954	18600	0.5463	25571232
0.693	0.6018	18800	0.5611	25847664
0.6779	0.6082	19000	0.5611	26125328
0.6462	0.6146	19200	0.5527	26404064
0.4534	0.6210	19400	0.5456	26677504
0.5426	0.6274	19600	0.5539	26952544
0.5724	0.6338	19800	0.5427	27226896
0.3026	0.6402	20000	0.5432	27501216
0.5014	0.6466	20200	0.5492	27776624
0.5602	0.6530	20400	0.5473	28051872
0.6304	0.6594	20600	0.5416	28325632
0.5525	0.6658	20800	0.5506	28598784
0.5865	0.6722	21000	0.5343	28874800
0.5382	0.6786	21200	0.5439	29151312
0.6467	0.6850	21400	0.5411	29425936
0.565	0.6914	21600	0.5407	29702784
0.6301	0.6978	21800	0.5329	29979824
0.4458	0.7042	22000	0.5339	30256128
0.452	0.7106	22200	0.5392	30528032
0.608	0.7170	22400	0.5369	30803904
0.7056	0.7234	22600	0.5329	31077632
0.5536	0.7298	22800	0.5303	31354544
0.594	0.7362	23000	0.5272	31626736
0.4026	0.7426	23200	0.5333	31901472
0.4775	0.7490	23400	0.5251	32179968
0.8036	0.7554	23600	0.5234	32457728
0.5097	0.7618	23800	0.5252	32732288
0.3575	0.7682	24000	0.5280	33007504
0.6269	0.7746	24200	0.5304	33281968
0.6274	0.7810	24400	0.5320	33558736
0.5539	0.7874	24600	0.5233	33830832
0.5876	0.7938	24800	0.5243	34104944
0.5314	0.8002	25000	0.5217	34381536
0.5158	0.8066	25200	0.5267	34654672
0.7053	0.8130	25400	0.5217	34931520
0.4934	0.8194	25600	0.5183	35206448
0.4797	0.8258	25800	0.5185	35482800
0.5238	0.8322	26000	0.5199	35756816
0.4021	0.8386	26200	0.5175	36031296
0.7042	0.8450	26400	0.5225	36307968
0.6731	0.8514	26600	0.5187	36580432
0.5761	0.8578	26800	0.5182	36855328
0.616	0.8642	27000	0.5133	37133072
0.3624	0.8706	27200	0.5141	37404464
0.5559	0.8770	27400	0.5117	37675456
0.4689	0.8834	27600	0.5123	37951616
0.5883	0.8898	27800	0.5095	38225840
0.4924	0.8962	28000	0.5094	38498736
0.5541	0.9026	28200	0.5092	38771760
0.4895	0.9090	28400	0.5091	39045824
0.3437	0.9154	28600	0.5097	39320736
0.4957	0.9218	28800	0.5113	39594816
0.5137	0.9282	29000	0.5090	39870432
0.5425	0.9346	29200	0.5091	40144672
0.4358	0.9410	29400	0.5062	40420752
0.4746	0.9474	29600	0.5064	40696672
0.3488	0.9538	29800	0.5055	40970096
0.5119	0.9602	30000	0.5056	41245904
0.6609	0.9666	30200	0.5052	41519232
0.5844	0.9730	30400	0.5048	41791520
0.4285	0.9795	30600	0.5039	42066928
0.56	0.9859	30800	0.5050	42339616
0.5263	0.9923	31000	0.5027	42616352
0.6078	0.9987	31200	0.5032	42892688
0.4005	1.0051	31400	0.5013	43167792
0.3178	1.0115	31600	0.5009	43444592
0.5995	1.0179	31800	0.4998	43719328
0.4926	1.0243	32000	0.4999	43994064
0.5995	1.0307	32200	0.4990	44269712
0.5196	1.0371	32400	0.4984	44545408
0.3068	1.0435	32600	0.4980	44819808
0.4889	1.0499	32800	0.4972	45097904
0.4359	1.0563	33000	0.4977	45376272
0.4718	1.0627	33200	0.4974	45647824
0.536	1.0691	33400	0.4979	45922032
0.5677	1.0755	33600	0.4972	46197840
0.4757	1.0819	33800	0.4974	46474848
0.5645	1.0883	34000	0.4970	46749824
0.5164	1.0947	34200	0.4962	47023856
0.4217	1.1011	34400	0.4973	47301520
0.5314	1.1075	34600	0.4966	47574864
0.7575	1.1139	34800	0.4967	47853888
0.3661	1.1203	35000	0.4956	48129792
0.5163	1.1267	35200	0.4956	48405024
0.3735	1.1331	35400	0.4951	48678592
0.3249	1.1395	35600	0.4948	48954048
0.3284	1.1459	35800	0.4949	49232480
0.6086	1.1523	36000	0.4946	49505040
0.4185	1.1587	36200	0.4948	49778864
0.4373	1.1651	36400	0.4945	50051632
0.3587	1.1715	36600	0.4946	50325888
0.456	1.1779	36800	0.4945	50601136
0.4731	1.1843	37000	0.4947	50876992
0.5226	1.1907	37200	0.4948	51153296
0.5659	1.1971	37400	0.4949	51427552
0.5782	1.2035	37600	0.4945	51707088
0.4875	1.2099	37800	0.4944	51981712
0.3913	1.2163	38000	0.4942	52254352
0.4959	1.2227	38200	0.4944	52529584
0.7156	1.2291	38400	0.4943	52803776
0.4096	1.2355	38600	0.4942	53078736
0.5082	1.2419	38800	0.4941	53352672
0.3781	1.2483	39000	0.4943	53628768
0.4386	1.2547	39200	0.4942	53905216
0.4696	1.2611	39400	0.4943	54178832
0.5154	1.2675	39600	0.4943	54454880
0.4841	1.2739	39800	0.4943	54727600
0.4098	1.2803	40000	0.4942	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