GSM8K-Binary_Llama-3.2-1B-fal8is8l

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: 0.9292
• Model Preparation Time: 0.0056
• Mdl: 3317.6903
• Accumulated Loss: 2299.6477
• Correct Preds: 1831.0
• Total Preds: 2475.0
• Accuracy: 0.7398
• Correct Gen Preds: 1492.0
• Gen Accuracy: 0.6028
• Correct Gen Preds 34192: 688.0
• Correct Preds 34192: 931.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.7784
• Gen Accuracy 34192: 0.5753
• Correct Gen Preds 41568: 796.0
• Correct Preds 41568: 900.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.7103
• Gen Accuracy 41568: 0.6283

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
1.3358	1.0	6	0.7779	0.0056	2777.7209	1925.3694	1292.0	2475.0	0.5220	8.0	0.0032	0.0	1091.0	1196.0	0.9122	0.0	0.0	201.0	1267.0	0.1586	0.0
0.7578	2.0	12	0.6802	0.0056	2428.6444	1683.4080	1649.0	2475.0	0.6663	7.0	0.0028	0.0	1004.0	1196.0	0.8395	0.0	0.0	645.0	1267.0	0.5091	0.0
1.201	3.0	18	0.7200	0.0056	2570.8615	1781.9854	1488.0	2475.0	0.6012	7.0	0.0028	0.0	1168.0	1196.0	0.9766	0.0	0.0	320.0	1267.0	0.2526	0.0
0.8365	4.0	24	0.6922	0.0056	2471.7795	1713.3070	1648.0	2475.0	0.6659	71.0	0.0287	0.0	1129.0	1196.0	0.9440	0.0	63.0	519.0	1267.0	0.4096	0.0497
0.4388	5.0	30	0.7937	0.0056	2834.1220	1964.4637	1674.0	2475.0	0.6764	117.0	0.0473	1.0	1118.0	1196.0	0.9348	0.0008	108.0	556.0	1267.0	0.4388	0.0852
0.1168	6.0	36	0.6699	0.0056	2392.0669	1658.0544	1758.0	2475.0	0.7103	242.0	0.0978	24.0	969.0	1196.0	0.8102	0.0201	210.0	789.0	1267.0	0.6227	0.1657
0.0328	7.0	42	0.9292	0.0056	3317.6903	2299.6477	1831.0	2475.0	0.7398	1492.0	0.6028	688.0	931.0	1196.0	0.7784	0.5753	796.0	900.0	1267.0	0.7103	0.6283
0.4624	8.0	48	1.4861	0.0056	5306.3463	3678.0790	1754.0	2475.0	0.7087	1681.0	0.6792	1018.0	1072.0	1196.0	0.8963	0.8512	656.0	682.0	1267.0	0.5383	0.5178
0.0041	9.0	54	1.3956	0.0056	4983.3753	3454.2125	1713.0	2475.0	0.6921	1328.0	0.5366	800.0	1080.0	1196.0	0.9030	0.6689	521.0	633.0	1267.0	0.4996	0.4112
0.0	10.0	60	1.2059	0.0056	4305.9186	2984.6353	1776.0	2475.0	0.7176	1744.0	0.7046	973.0	996.0	1196.0	0.8328	0.8135	764.0	780.0	1267.0	0.6156	0.6030
0.0001	11.0	66	1.2530	0.0056	4474.1803	3101.2654	1797.0	2475.0	0.7261	1797.0	0.7261	959.0	961.0	1196.0	0.8035	0.8018	830.0	836.0	1267.0	0.6598	0.6551
0.0019	12.0	72	1.2914	0.0056	4611.0049	3196.1051	1806.0	2475.0	0.7297	1812.0	0.7321	831.0	832.0	1196.0	0.6957	0.6948	973.0	974.0	1267.0	0.7687	0.7680
0.0	13.0	78	1.4823	0.0056	5292.8256	3668.7071	1752.0	2475.0	0.7079	1756.0	0.7095	1028.0	1029.0	1196.0	0.8604	0.8595	720.0	723.0	1267.0	0.5706	0.5683
0.9048	14.0	84	1.5439	0.0056	5512.8836	3821.2397	1744.0	2475.0	0.7046	1748.0	0.7063	1041.0	1042.0	1196.0	0.8712	0.8704	699.0	702.0	1267.0	0.5541	0.5517
0.4524	15.0	90	1.5327	0.0056	5472.8181	3793.4684	1741.0	2475.0	0.7034	1744.0	0.7046	1029.0	1031.0	1196.0	0.8620	0.8604	707.0	710.0	1267.0	0.5604	0.5580
0.0001	16.0	96	1.5159	0.0056	5412.8149	3751.8774	1750.0	2475.0	0.7071	1752.0	0.7079	1027.0	1029.0	1196.0	0.8604	0.8587	717.0	721.0	1267.0	0.5691	0.5659
0.0	17.0	102	1.4979	0.0056	5348.5086	3707.3037	1760.0	2475.0	0.7111	1762.0	0.7119	1022.0	1024.0	1196.0	0.8562	0.8545	732.0	736.0	1267.0	0.5809	0.5777
0.4524	18.0	108	1.4874	0.0056	5311.1411	3681.4025	1760.0	2475.0	0.7111	1764.0	0.7127	1018.0	1018.0	1196.0	0.8512	0.8512	738.0	742.0	1267.0	0.5856	0.5825
0.0	19.0	114	1.4810	0.0056	5288.0565	3665.4015	1771.0	2475.0	0.7156	1774.0	0.7168	1018.0	1018.0	1196.0	0.8512	0.8512	748.0	753.0	1267.0	0.5943	0.5904
0.0	20.0	120	1.4760	0.0056	5270.4369	3653.1885	1773.0	2475.0	0.7164	1776.0	0.7176	1014.0	1014.0	1196.0	0.8478	0.8478	754.0	759.0	1267.0	0.5991	0.5951
0.0	21.0	126	1.4728	0.0056	5258.7289	3645.0731	1773.0	2475.0	0.7164	1776.0	0.7176	1013.0	1013.0	1196.0	0.8470	0.8470	755.0	760.0	1267.0	0.5998	0.5959
0.0	22.0	132	1.4693	0.0056	5246.4381	3636.5537	1777.0	2475.0	0.7180	1779.0	0.7188	1013.0	1014.0	1196.0	0.8478	0.8470	758.0	763.0	1267.0	0.6022	0.5983
0.9048	23.0	138	1.4666	0.0056	5236.6019	3629.7358	1774.0	2475.0	0.7168	1776.0	0.7176	1012.0	1013.0	1196.0	0.8470	0.8462	756.0	761.0	1267.0	0.6006	0.5967
0.4524	24.0	144	1.4641	0.0056	5227.7933	3623.6302	1779.0	2475.0	0.7188	1782.0	0.72	1014.0	1014.0	1196.0	0.8478	0.8478	760.0	765.0	1267.0	0.6038	0.5998
0.0	25.0	150	1.4629	0.0056	5223.5871	3620.7147	1777.0	2475.0	0.7180	1780.0	0.7192	1012.0	1012.0	1196.0	0.8462	0.8462	760.0	765.0	1267.0	0.6038	0.5998
0.0	26.0	156	1.4613	0.0056	5217.9147	3616.7828	1781.0	2475.0	0.7196	1784.0	0.7208	1013.0	1013.0	1196.0	0.8470	0.8470	763.0	768.0	1267.0	0.6062	0.6022
0.0	27.0	162	1.4584	0.0056	5207.4129	3609.5036	1781.0	2475.0	0.7196	1784.0	0.7208	1011.0	1011.0	1196.0	0.8453	0.8453	765.0	770.0	1267.0	0.6077	0.6038
0.0	28.0	168	1.4583	0.0056	5207.2841	3609.4143	1779.0	2475.0	0.7188	1782.0	0.72	1009.0	1009.0	1196.0	0.8436	0.8436	765.0	770.0	1267.0	0.6077	0.6038
0.4524	29.0	174	1.4559	0.0056	5198.4690	3603.3041	1786.0	2475.0	0.7216	1790.0	0.7232	1012.0	1012.0	1196.0	0.8462	0.8462	770.0	774.0	1267.0	0.6109	0.6077
0.0	30.0	180	1.4550	0.0056	5195.2049	3601.0416	1785.0	2475.0	0.7212	1788.0	0.7224	1011.0	1011.0	1196.0	0.8453	0.8453	769.0	774.0	1267.0	0.6109	0.6069
0.0	31.0	186	1.4533	0.0056	5189.0929	3596.8051	1783.0	2475.0	0.7204	1787.0	0.7220	1011.0	1011.0	1196.0	0.8453	0.8453	768.0	772.0	1267.0	0.6093	0.6062
0.4524	32.0	192	1.4546	0.0056	5193.8924	3600.1319	1785.0	2475.0	0.7212	1789.0	0.7228	1009.0	1009.0	1196.0	0.8436	0.8436	772.0	776.0	1267.0	0.6125	0.6093
0.4524	33.0	198	1.4522	0.0056	5185.2381	3594.1332	1786.0	2475.0	0.7216	1789.0	0.7228	1011.0	1011.0	1196.0	0.8453	0.8453	770.0	775.0	1267.0	0.6117	0.6077
0.0	34.0	204	1.4512	0.0056	5181.7853	3591.7399	1785.0	2475.0	0.7212	1788.0	0.7224	1008.0	1008.0	1196.0	0.8428	0.8428	772.0	777.0	1267.0	0.6133	0.6093
0.4524	35.0	210	1.4503	0.0056	5178.6494	3589.5662	1786.0	2475.0	0.7216	1789.0	0.7228	1009.0	1009.0	1196.0	0.8436	0.8436	772.0	777.0	1267.0	0.6133	0.6093
0.0	36.0	216	1.4496	0.0056	5175.8705	3587.6401	1788.0	2475.0	0.7224	1792.0	0.7240	1007.0	1007.0	1196.0	0.8420	0.8420	777.0	781.0	1267.0	0.6164	0.6133
0.4524	37.0	222	1.4509	0.0056	5180.7493	3591.0218	1783.0	2475.0	0.7204	1787.0	0.7220	1005.0	1005.0	1196.0	0.8403	0.8403	774.0	778.0	1267.0	0.6140	0.6109

Framework versions

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