train_record_1745950246

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: 0.7299
• 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
2.8233	0.0064	200	4.0758	277264
1.6949	0.0128	400	1.6427	548976
1.4877	0.0192	600	1.3227	826016
1.3337	0.0256	800	1.2206	1099968
1.2957	0.0320	1000	1.1480	1374672
1.0295	0.0384	1200	1.1003	1647936
1.0462	0.0448	1400	1.0607	1921648
0.8464	0.0512	1600	1.0352	2194448
0.8272	0.0576	1800	1.0096	2472048
1.1763	0.0640	2000	0.9865	2746752
1.3931	0.0704	2200	0.9697	3020144
1.2572	0.0768	2400	0.9548	3296624
1.0349	0.0832	2600	0.9399	3571808
1.1218	0.0896	2800	0.9313	3847184
0.8946	0.0960	3000	0.9185	4121024
0.7198	0.1024	3200	0.9094	4396880
0.8046	0.1088	3400	0.9014	4671152
0.8053	0.1152	3600	0.8916	4950800
0.9251	0.1216	3800	0.8846	5228512
0.845	0.1280	4000	0.8784	5504608
0.9929	0.1344	4200	0.8734	5778176
0.7947	0.1408	4400	0.8657	6055712
0.6828	0.1472	4600	0.8624	6331680
0.7723	0.1536	4800	0.8569	6604544
0.8328	0.1600	5000	0.8511	6882256
0.8425	0.1664	5200	0.8481	7159072
0.7358	0.1728	5400	0.8421	7433136
0.9727	0.1792	5600	0.8378	7707776
0.6727	0.1856	5800	0.8367	7985472
0.7569	0.1920	6000	0.8327	8259552
1.0037	0.1985	6200	0.8338	8535952
0.661	0.2049	6400	0.8256	8809968
0.9558	0.2113	6600	0.8226	9084016
0.869	0.2177	6800	0.8198	9357456
0.6787	0.2241	7000	0.8197	9630608
0.6022	0.2305	7200	0.8149	9907888
0.6105	0.2369	7400	0.8124	10182048
1.0618	0.2433	7600	0.8104	10458544
0.7117	0.2497	7800	0.8075	10736144
0.8245	0.2561	8000	0.8047	11010512
0.6891	0.2625	8200	0.8023	11284128
0.8266	0.2689	8400	0.8010	11556816
0.9607	0.2753	8600	0.7989	11828816
0.6199	0.2817	8800	0.7967	12104176
0.733	0.2881	9000	0.7948	12378784
0.7747	0.2945	9200	0.7927	12654368
0.6677	0.3009	9400	0.7914	12927088
0.9276	0.3073	9600	0.7888	13199552
0.8887	0.3137	9800	0.7877	13473952
0.7798	0.3201	10000	0.7864	13750288
0.7029	0.3265	10200	0.7865	14025248
0.7353	0.3329	10400	0.7841	14300160
0.6306	0.3393	10600	0.7833	14577760
0.7592	0.3457	10800	0.7805	14851280
0.588	0.3521	11000	0.7808	15125104
0.7482	0.3585	11200	0.7781	15398624
0.9051	0.3649	11400	0.7771	15672384
0.8379	0.3713	11600	0.7754	15946384
0.6137	0.3777	11800	0.7736	16220112
0.7876	0.3841	12000	0.7733	16493920
0.7736	0.3905	12200	0.7744	16771376
0.877	0.3969	12400	0.7717	17046656
0.6752	0.4033	12600	0.7706	17318272
0.6949	0.4097	12800	0.7698	17591696
0.7197	0.4161	13000	0.7689	17864256
0.6706	0.4225	13200	0.7704	18137984
0.8119	0.4289	13400	0.7684	18413504
0.8481	0.4353	13600	0.7665	18690528
0.8423	0.4417	13800	0.7652	18966352
0.8398	0.4481	14000	0.7640	19242160
0.7095	0.4545	14200	0.7629	19518832
0.7228	0.4609	14400	0.7624	19795920
0.4926	0.4673	14600	0.7622	20073168
0.4827	0.4737	14800	0.7614	20349056
0.9413	0.4801	15000	0.7603	20622896
0.83	0.4865	15200	0.7598	20896768
0.5562	0.4929	15400	0.7583	21171376
0.7659	0.4993	15600	0.7580	21447568
0.9216	0.5057	15800	0.7575	21722256
0.7743	0.5121	16000	0.7572	21998320
0.6469	0.5185	16200	0.7558	22273616
0.7036	0.5249	16400	0.7551	22549280
0.8155	0.5313	16600	0.7546	22823984
0.7568	0.5377	16800	0.7551	23098384
0.7415	0.5441	17000	0.7532	23371136
0.5823	0.5505	17200	0.7527	23647856
0.6466	0.5569	17400	0.7518	23921008
0.7807	0.5633	17600	0.7514	24194480
0.7174	0.5697	17800	0.7505	24469312
0.7576	0.5761	18000	0.7501	24743360
0.961	0.5825	18200	0.7497	25020352
0.9696	0.5890	18400	0.7490	25295920
0.6775	0.5954	18600	0.7484	25571232
0.8449	0.6018	18800	0.7478	25847664
0.8842	0.6082	19000	0.7478	26125328
0.7932	0.6146	19200	0.7467	26404064
0.6471	0.6210	19400	0.7466	26677504
0.8983	0.6274	19600	0.7469	26952544
0.7124	0.6338	19800	0.7465	27226896
0.7017	0.6402	20000	0.7465	27501216
0.6379	0.6466	20200	0.7448	27776624
0.6725	0.6530	20400	0.7443	28051872
0.6149	0.6594	20600	0.7431	28325632
0.9104	0.6658	20800	0.7431	28598784
0.7189	0.6722	21000	0.7426	28874800
0.6754	0.6786	21200	0.7428	29151312
1.0405	0.6850	21400	0.7422	29425936
0.8179	0.6914	21600	0.7416	29702784
0.9606	0.6978	21800	0.7411	29979824
0.7739	0.7042	22000	0.7404	30256128
0.6248	0.7106	22200	0.7403	30528032
0.7647	0.7170	22400	0.7410	30803904
0.828	0.7234	22600	0.7406	31077632
1.0034	0.7298	22800	0.7403	31354544
0.6809	0.7362	23000	0.7400	31626736
0.6563	0.7426	23200	0.7396	31901472
0.7809	0.7490	23400	0.7391	32179968
0.8411	0.7554	23600	0.7388	32457728
0.7451	0.7618	23800	0.7382	32732288
0.7342	0.7682	24000	0.7381	33007504
0.7153	0.7746	24200	0.7375	33281968
0.7706	0.7810	24400	0.7381	33558736
0.7155	0.7874	24600	0.7376	33830832
0.9201	0.7938	24800	0.7376	34104944
0.7624	0.8002	25000	0.7368	34381536
0.6255	0.8066	25200	0.7362	34654672
0.7843	0.8130	25400	0.7355	34931520
0.6743	0.8194	25600	0.7354	35206448
0.632	0.8258	25800	0.7356	35482800
0.8752	0.8322	26000	0.7352	35756816
0.6129	0.8386	26200	0.7356	36031296
0.7504	0.8450	26400	0.7348	36307968
0.9538	0.8514	26600	0.7345	36580432
0.7165	0.8578	26800	0.7340	36855328
0.8094	0.8642	27000	0.7339	37133072
0.7163	0.8706	27200	0.7340	37404464
0.7094	0.8770	27400	0.7335	37675456
0.6543	0.8834	27600	0.7332	37951616
0.6482	0.8898	27800	0.7332	38225840
0.7161	0.8962	28000	0.7330	38498736
0.6364	0.9026	28200	0.7328	38771760
0.7501	0.9090	28400	0.7328	39045824
0.505	0.9154	28600	0.7328	39320736
0.6496	0.9218	28800	0.7324	39594816
0.7856	0.9282	29000	0.7326	39870432
0.6598	0.9346	29200	0.7323	40144672
0.7067	0.9410	29400	0.7323	40420752
0.7768	0.9474	29600	0.7318	40696672
0.4495	0.9538	29800	0.7317	40970096
0.6259	0.9602	30000	0.7321	41245904
0.9303	0.9666	30200	0.7320	41519232
1.0491	0.9730	30400	0.7319	41791520
0.6816	0.9795	30600	0.7319	42066928
0.8694	0.9859	30800	0.7326	42339616
0.9029	0.9923	31000	0.7318	42616352
0.8713	0.9987	31200	0.7324	42892688
0.4775	1.0051	31400	0.7324	43167792
0.7046	1.0115	31600	0.7319	43444592
0.7093	1.0179	31800	0.7318	43719328
0.58	1.0243	32000	0.7312	43994064
0.7032	1.0307	32200	0.7314	44269712
0.7959	1.0371	32400	0.7309	44545408
0.5933	1.0435	32600	0.7311	44819808
0.6621	1.0499	32800	0.7309	45097904
0.7677	1.0563	33000	0.7308	45376272
0.7354	1.0627	33200	0.7307	45647824
0.7796	1.0691	33400	0.7308	45922032
0.7574	1.0755	33600	0.7306	46197840
0.8987	1.0819	33800	0.7303	46474848
0.6623	1.0883	34000	0.7307	46749824
0.8771	1.0947	34200	0.7308	47023856
0.6231	1.1011	34400	0.7305	47301520
0.6705	1.1075	34600	0.7306	47574864
1.0773	1.1139	34800	0.7307	47853888
0.5722	1.1203	35000	0.7305	48129792
0.7991	1.1267	35200	0.7307	48405024
0.7533	1.1331	35400	0.7306	48678592
0.7164	1.1395	35600	0.7302	48954048
0.6525	1.1459	35800	0.7302	49232480
0.7326	1.1523	36000	0.7303	49505040
0.6382	1.1587	36200	0.7303	49778864
0.7743	1.1651	36400	0.7299	50051632
0.5618	1.1715	36600	0.7302	50325888
0.5741	1.1779	36800	0.7303	50601136
0.7073	1.1843	37000	0.7300	50876992
0.8427	1.1907	37200	0.7301	51153296
0.9195	1.1971	37400	0.7300	51427552
0.8146	1.2035	37600	0.7300	51707088
0.682	1.2099	37800	0.7302	51981712
0.6585	1.2163	38000	0.7302	52254352
0.7149	1.2227	38200	0.7300	52529584
0.892	1.2291	38400	0.7299	52803776
0.626	1.2355	38600	0.7300	53078736
0.6822	1.2419	38800	0.7300	53352672
0.5693	1.2483	39000	0.7302	53628768
0.7027	1.2547	39200	0.7301	53905216
0.838	1.2611	39400	0.7300	54178832
0.5893	1.2675	39600	0.7300	54454880
0.818	1.2739	39800	0.7300	54727600
0.6806	1.2803	40000	0.7300	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