GSM8K-Binary_Llama-3.2-1B-gups5p1g

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7591
• Model Preparation Time: 0.0056
• Mdl: 6281.2318
• Accumulated Loss: 4353.8181
• Correct Preds: 1735.0
• Total Preds: 2475.0
• Accuracy: 0.7010
• Correct Gen Preds: 1587.0
• Gen Accuracy: 0.6412
• Correct Gen Preds 34192: 948.0
• Correct Preds 34192: 1022.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8545
• Gen Accuracy 34192: 0.7926
• Correct Gen Preds 41568: 631.0
• Correct Preds 41568: 713.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.5627
• Gen Accuracy 41568: 0.4980

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Mdl	Accumulated Loss	Correct Preds	Total Preds	Accuracy	Correct Gen Preds	Gen Accuracy	Correct Gen Preds 34192	Correct Preds 34192	Total Labels 34192	Accuracy 34192	Gen Accuracy 34192	Correct Gen Preds 41568	Correct Preds 41568	Total Labels 41568	Accuracy 41568	Gen Accuracy 41568
No log	0	0	1.4656	0.0056	5233.1723	3627.3586	1196.0	2475.0	0.4832	1204.0	0.4865	1196.0	1196.0	1196.0	1.0	1.0	0.0	0.0	1267.0	0.0	0.0
1.1136	1.0	2	2.4276	0.0056	8668.2571	6008.3780	1266.0	2475.0	0.5115	1271.0	0.5135	0.0	0.0	1196.0	0.0	0.0	1263.0	1266.0	1267.0	0.9992	0.9968
0.2026	2.0	4	2.7999	0.0056	9997.6648	6929.8532	1267.0	2475.0	0.5119	1275.0	0.5152	0.0	0.0	1196.0	0.0	0.0	1266.0	1267.0	1267.0	1.0	0.9992
0.804	3.0	6	1.5898	0.0056	5676.6872	3934.7797	1196.0	2475.0	0.4832	953.0	0.3851	945.0	1196.0	1196.0	1.0	0.7901	0.0	0.0	1267.0	0.0	0.0
0.7582	4.0	8	0.7570	0.0056	2702.9597	1873.5489	1185.0	2475.0	0.4788	8.0	0.0032	0.0	797.0	1196.0	0.6664	0.0	0.0	388.0	1267.0	0.3062	0.0
0.7096	5.0	10	0.7636	0.0056	2726.5581	1889.9061	1246.0	2475.0	0.5034	480.0	0.1939	0.0	3.0	1196.0	0.0025	0.0	472.0	1243.0	1267.0	0.9811	0.3725
0.584	6.0	12	0.7512	0.0056	2682.1163	1859.1013	1247.0	2475.0	0.5038	1085.0	0.4384	1057.0	1188.0	1196.0	0.9933	0.8838	20.0	59.0	1267.0	0.0466	0.0158
1.2302	7.0	14	0.7259	0.0056	2591.8945	1796.5643	1462.0	2475.0	0.5907	366.0	0.1479	266.0	1085.0	1196.0	0.9072	0.2224	92.0	377.0	1267.0	0.2976	0.0726
0.5839	8.0	16	1.0453	0.0056	3732.5655	2587.2173	1230.0	2475.0	0.4970	1193.0	0.4820	1185.0	1196.0	1196.0	1.0	0.9908	0.0	34.0	1267.0	0.0268	0.0
0.3271	9.0	18	0.6844	0.0056	2443.8585	1693.9537	1615.0	2475.0	0.6525	656.0	0.2651	285.0	713.0	1196.0	0.5962	0.2383	363.0	902.0	1267.0	0.7119	0.2865
0.268	10.0	20	0.8349	0.0056	2981.1079	2066.3465	1361.0	2475.0	0.5499	1109.0	0.4481	1058.0	1155.0	1196.0	0.9657	0.8846	43.0	206.0	1267.0	0.1626	0.0339
0.6073	11.0	22	0.7529	0.0056	2688.4658	1863.5025	1398.0	2475.0	0.5648	618.0	0.2497	42.0	176.0	1196.0	0.1472	0.0351	568.0	1222.0	1267.0	0.9645	0.4483
0.4368	12.0	24	0.6971	0.0056	2488.9442	1725.2046	1533.0	2475.0	0.6194	867.0	0.3503	731.0	1113.0	1196.0	0.9306	0.6112	128.0	420.0	1267.0	0.3315	0.1010
0.3446	13.0	26	0.6723	0.0056	2400.6893	1664.0310	1661.0	2475.0	0.6711	635.0	0.2566	368.0	867.0	1196.0	0.7249	0.3077	259.0	794.0	1267.0	0.6267	0.2044
0.2391	14.0	28	0.7026	0.0056	2508.8014	1738.9686	1666.0	2475.0	0.6731	311.0	0.1257	75.0	662.0	1196.0	0.5535	0.0627	228.0	1004.0	1267.0	0.7924	0.1800
0.0074	15.0	30	1.0296	0.0056	3676.4104	2548.2935	1536.0	2475.0	0.6206	580.0	0.2343	495.0	1121.0	1196.0	0.9373	0.4139	77.0	415.0	1267.0	0.3275	0.0608
0.0596	16.0	32	0.8138	0.0056	2905.7954	2014.1439	1725.0	2475.0	0.6970	353.0	0.1426	272.0	1014.0	1196.0	0.8478	0.2274	73.0	711.0	1267.0	0.5612	0.0576
0.0001	17.0	34	1.0267	0.0056	3666.0603	2541.1194	1680.0	2475.0	0.6788	428.0	0.1729	355.0	1041.0	1196.0	0.8704	0.2968	65.0	639.0	1267.0	0.5043	0.0513
0.0427	18.0	36	1.8264	0.0056	6521.4362	4520.3151	1575.0	2475.0	0.6364	747.0	0.3018	645.0	1106.0	1196.0	0.9247	0.5393	94.0	469.0	1267.0	0.3702	0.0742
0.0008	19.0	38	1.6238	0.0056	5798.0140	4018.8771	1692.0	2475.0	0.6836	1043.0	0.4214	743.0	1038.0	1196.0	0.8679	0.6212	292.0	654.0	1267.0	0.5162	0.2305
0.0001	20.0	40	2.0709	0.0056	7394.6695	5125.5943	1662.0	2475.0	0.6715	1331.0	0.5378	962.0	1091.0	1196.0	0.9122	0.8043	361.0	571.0	1267.0	0.4507	0.2849
0.0001	21.0	42	1.7667	0.0056	6308.1344	4372.4656	1721.0	2475.0	0.6954	1463.0	0.5911	918.0	1043.0	1196.0	0.8721	0.7676	537.0	678.0	1267.0	0.5351	0.4238
0.0001	22.0	44	1.7591	0.0056	6281.2318	4353.8181	1735.0	2475.0	0.7010	1587.0	0.6412	948.0	1022.0	1196.0	0.8545	0.7926	631.0	713.0	1267.0	0.5627	0.4980
0.0002	23.0	46	1.9850	0.0056	7087.6303	4912.7710	1723.0	2475.0	0.6962	1636.0	0.6610	1008.0	1045.0	1196.0	0.8737	0.8428	620.0	678.0	1267.0	0.5351	0.4893
0.0	24.0	48	2.3210	0.0056	8287.5460	5744.4892	1695.0	2475.0	0.6848	1637.0	0.6614	1050.0	1075.0	1196.0	0.8988	0.8779	578.0	620.0	1267.0	0.4893	0.4562
0.0	25.0	50	2.5588	0.0056	9136.5495	6332.9735	1681.0	2475.0	0.6792	1647.0	0.6655	1079.0	1095.0	1196.0	0.9156	0.9022	560.0	586.0	1267.0	0.4625	0.4420
0.0	26.0	52	2.7194	0.0056	9710.0766	6730.5122	1668.0	2475.0	0.6739	1646.0	0.6651	1094.0	1100.0	1196.0	0.9197	0.9147	544.0	568.0	1267.0	0.4483	0.4294
0.0	27.0	54	2.8232	0.0056	10080.7412	6987.4374	1654.0	2475.0	0.6683	1637.0	0.6614	1096.0	1101.0	1196.0	0.9206	0.9164	533.0	553.0	1267.0	0.4365	0.4207
0.0001	28.0	56	2.8833	0.0056	10295.4086	7136.2334	1648.0	2475.0	0.6659	1635.0	0.6606	1098.0	1101.0	1196.0	0.9206	0.9181	528.0	547.0	1267.0	0.4317	0.4167
0.0	29.0	58	2.9126	0.0056	10399.8231	7208.6081	1649.0	2475.0	0.6663	1633.0	0.6598	1099.0	1104.0	1196.0	0.9231	0.9189	525.0	545.0	1267.0	0.4301	0.4144
0.0	30.0	60	2.9375	0.0056	10488.7908	7270.2758	1648.0	2475.0	0.6659	1637.0	0.6614	1101.0	1105.0	1196.0	0.9239	0.9206	527.0	543.0	1267.0	0.4286	0.4159
0.0	31.0	62	2.9496	0.0056	10531.9552	7300.1950	1647.0	2475.0	0.6655	1636.0	0.6610	1100.0	1104.0	1196.0	0.9231	0.9197	527.0	543.0	1267.0	0.4286	0.4159
0.0	32.0	64	2.9590	0.0056	10565.7761	7323.6379	1652.0	2475.0	0.6675	1640.0	0.6626	1104.0	1106.0	1196.0	0.9247	0.9231	527.0	546.0	1267.0	0.4309	0.4159
0.0	33.0	66	2.9643	0.0056	10584.5428	7336.6460	1647.0	2475.0	0.6655	1640.0	0.6626	1102.0	1105.0	1196.0	0.9239	0.9214	530.0	542.0	1267.0	0.4278	0.4183
0.0	34.0	68	2.9646	0.0056	10585.7729	7337.4987	1651.0	2475.0	0.6671	1641.0	0.6630	1102.0	1106.0	1196.0	0.9247	0.9214	531.0	545.0	1267.0	0.4301	0.4191
0.0	35.0	70	2.9645	0.0056	10585.3148	7337.1811	1650.0	2475.0	0.6667	1639.0	0.6622	1101.0	1106.0	1196.0	0.9247	0.9206	530.0	544.0	1267.0	0.4294	0.4183
0.0	36.0	72	2.9663	0.0056	10591.7568	7341.6464	1653.0	2475.0	0.6679	1644.0	0.6642	1101.0	1105.0	1196.0	0.9239	0.9206	534.0	548.0	1267.0	0.4325	0.4215
0.0	37.0	74	2.9617	0.0056	10575.2913	7330.2333	1650.0	2475.0	0.6667	1642.0	0.6634	1101.0	1103.0	1196.0	0.9222	0.9206	533.0	547.0	1267.0	0.4317	0.4207
0.0	38.0	76	2.9582	0.0056	10562.7169	7321.5174	1655.0	2475.0	0.6687	1649.0	0.6663	1101.0	1104.0	1196.0	0.9231	0.9206	539.0	551.0	1267.0	0.4349	0.4254
0.0	39.0	78	2.9564	0.0056	10556.4163	7317.1502	1656.0	2475.0	0.6691	1648.0	0.6659	1100.0	1104.0	1196.0	0.9231	0.9197	539.0	552.0	1267.0	0.4357	0.4254
0.0	40.0	80	2.9507	0.0056	10536.1381	7303.0944	1654.0	2475.0	0.6683	1648.0	0.6659	1099.0	1102.0	1196.0	0.9214	0.9189	541.0	552.0	1267.0	0.4357	0.4270
0.0	41.0	82	2.9500	0.0056	10533.3466	7301.1595	1652.0	2475.0	0.6675	1643.0	0.6638	1097.0	1100.0	1196.0	0.9197	0.9172	537.0	552.0	1267.0	0.4357	0.4238
0.0	42.0	84	2.9491	0.0056	10530.1332	7298.9322	1658.0	2475.0	0.6699	1652.0	0.6675	1100.0	1103.0	1196.0	0.9222	0.9197	543.0	555.0	1267.0	0.4380	0.4286
0.0	43.0	86	2.9488	0.0056	10529.0906	7298.2094	1659.0	2475.0	0.6703	1651.0	0.6671	1099.0	1104.0	1196.0	0.9231	0.9189	544.0	555.0	1267.0	0.4380	0.4294
0.0	44.0	88	2.9497	0.0056	10532.3216	7300.4490	1656.0	2475.0	0.6691	1651.0	0.6671	1099.0	1101.0	1196.0	0.9206	0.9189	543.0	555.0	1267.0	0.4380	0.4286
0.0	45.0	90	2.9481	0.0056	10526.6856	7296.5424	1661.0	2475.0	0.6711	1653.0	0.6679	1100.0	1104.0	1196.0	0.9231	0.9197	545.0	557.0	1267.0	0.4396	0.4301
0.0	46.0	92	2.9468	0.0056	10522.2084	7293.4391	1658.0	2475.0	0.6699	1651.0	0.6671	1099.0	1101.0	1196.0	0.9206	0.9189	543.0	557.0	1267.0	0.4396	0.4286
0.0	47.0	94	2.9431	0.0056	10508.9090	7284.2206	1655.0	2475.0	0.6687	1649.0	0.6663	1099.0	1101.0	1196.0	0.9206	0.9189	541.0	554.0	1267.0	0.4373	0.4270
0.0	48.0	96	2.9471	0.0056	10523.2683	7294.1737	1659.0	2475.0	0.6703	1651.0	0.6671	1101.0	1103.0	1196.0	0.9222	0.9206	541.0	556.0	1267.0	0.4388	0.4270
0.0	49.0	98	2.9479	0.0056	10525.8801	7295.9841	1655.0	2475.0	0.6687	1651.0	0.6671	1099.0	1102.0	1196.0	0.9214	0.9189	543.0	553.0	1267.0	0.4365	0.4286
0.0	50.0	100	2.9473	0.0056	10523.6648	7294.4486	1655.0	2475.0	0.6687	1648.0	0.6659	1098.0	1101.0	1196.0	0.9206	0.9181	541.0	554.0	1267.0	0.4373	0.4270
0.0	51.0	102	2.9465	0.0056	10520.8765	7292.5159	1653.0	2475.0	0.6679	1647.0	0.6655	1099.0	1101.0	1196.0	0.9206	0.9189	540.0	552.0	1267.0	0.4357	0.4262
0.0	52.0	104	2.9463	0.0056	10520.2917	7292.1105	1654.0	2475.0	0.6683	1649.0	0.6663	1099.0	1102.0	1196.0	0.9214	0.9189	541.0	552.0	1267.0	0.4357	0.4270

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1