GSM8K-Binary_Llama-3.2-1B-thbeu27y

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: 1.6734
• Model Preparation Time: 0.0026
• Mdl: 5975.0439
• Accumulated Loss: 4141.5848
• Correct Preds: 2006.0
• Total Preds: 2475.0
• Accuracy: 0.8105
• Correct Gen Preds: 2008.0
• Gen Accuracy: 0.8113
• Correct Gen Preds 34192: 1008.0
• Correct Preds 34192: 1011.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8453
• Gen Accuracy 34192: 0.8428
• Correct Gen Preds 41568: 992.0
• Correct Preds 41568: 995.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.7853
• Gen Accuracy 41568: 0.7830

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	0.0026	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
0.5402	1.0	52	0.5851	0.0026	2089.1467	1448.0861	1830.0	2475.0	0.7394	1830.0	0.7394	707.0	711.0	1196.0	0.5945	0.5911	1115.0	1119.0	1267.0	0.8832	0.8800
0.6372	2.0	104	0.4997	0.0026	1784.3730	1236.8331	1971.0	2475.0	0.7964	1778.0	0.7184	907.0	969.0	1196.0	0.8102	0.7584	862.0	1002.0	1267.0	0.7908	0.6803
0.4523	3.0	156	0.4722	0.0026	1686.1189	1168.7286	1996.0	2475.0	0.8065	1420.0	0.5737	732.0	979.0	1196.0	0.8186	0.6120	681.0	1017.0	1267.0	0.8027	0.5375
0.1558	4.0	208	0.6308	0.0026	2252.3394	1561.2027	1920.0	2475.0	0.7758	1016.0	0.4105	737.0	1129.0	1196.0	0.9440	0.6162	271.0	791.0	1267.0	0.6243	0.2139
0.2356	5.0	260	0.6758	0.0026	2413.1342	1672.6572	1992.0	2475.0	0.8048	1324.0	0.5349	516.0	957.0	1196.0	0.8002	0.4314	800.0	1035.0	1267.0	0.8169	0.6314
0.0194	6.0	312	0.9312	0.0026	3325.0850	2304.7733	1940.0	2475.0	0.7838	1685.0	0.6808	961.0	1108.0	1196.0	0.9264	0.8035	715.0	832.0	1267.0	0.6567	0.5643
0.028	7.0	364	1.1226	0.0026	4008.2597	2778.3139	1998.0	2475.0	0.8073	1949.0	0.7875	983.0	1011.0	1196.0	0.8453	0.8219	956.0	987.0	1267.0	0.7790	0.7545
0.3	8.0	416	1.1704	0.0026	4178.9701	2896.6413	1929.0	2475.0	0.7794	1915.0	0.7737	813.0	827.0	1196.0	0.6915	0.6798	1092.0	1102.0	1267.0	0.8698	0.8619
0.5882	9.0	468	1.5126	0.0026	5400.9251	3743.6360	1996.0	2475.0	0.8065	1983.0	0.8012	1035.0	1046.0	1196.0	0.8746	0.8654	940.0	950.0	1267.0	0.7498	0.7419
0.0	10.0	520	1.6734	0.0026	5975.0439	4141.5848	2006.0	2475.0	0.8105	2008.0	0.8113	1008.0	1011.0	1196.0	0.8453	0.8428	992.0	995.0	1267.0	0.7853	0.7830
0.5881	11.0	572	1.7454	0.0026	6232.2057	4319.8358	1999.0	2475.0	0.8077	1999.0	0.8077	1024.0	1028.0	1196.0	0.8595	0.8562	967.0	971.0	1267.0	0.7664	0.7632
0.0	12.0	624	1.7462	0.0026	6235.0887	4321.8342	1997.0	2475.0	0.8069	1999.0	0.8077	1018.0	1021.0	1196.0	0.8537	0.8512	973.0	976.0	1267.0	0.7703	0.7680
0.5881	13.0	676	1.7479	0.0026	6241.0599	4325.9731	1998.0	2475.0	0.8073	2000.0	0.8081	1020.0	1023.0	1196.0	0.8554	0.8528	972.0	975.0	1267.0	0.7695	0.7672
0.0	14.0	728	1.7475	0.0026	6239.8590	4325.1407	1998.0	2475.0	0.8073	2001.0	0.8085	1020.0	1023.0	1196.0	0.8554	0.8528	973.0	975.0	1267.0	0.7695	0.7680
0.0	15.0	780	1.7505	0.0026	6250.3925	4332.4419	1998.0	2475.0	0.8073	2000.0	0.8081	1021.0	1025.0	1196.0	0.8570	0.8537	971.0	973.0	1267.0	0.7680	0.7664
0.0	16.0	832	1.7500	0.0026	6248.8051	4331.3416	1998.0	2475.0	0.8073	1999.0	0.8077	1020.0	1023.0	1196.0	0.8554	0.8528	971.0	975.0	1267.0	0.7695	0.7664
0.0	17.0	884	1.7511	0.0026	6252.4822	4333.8904	1999.0	2475.0	0.8077	2002.0	0.8089	1021.0	1024.0	1196.0	0.8562	0.8537	973.0	975.0	1267.0	0.7695	0.7680
0.0	18.0	936	1.7545	0.0026	6264.7927	4342.4234	1999.0	2475.0	0.8077	2000.0	0.8081	1020.0	1024.0	1196.0	0.8562	0.8528	972.0	975.0	1267.0	0.7695	0.7672
0.0	19.0	988	1.7568	0.0026	6272.8945	4348.0391	1997.0	2475.0	0.8069	1999.0	0.8077	1019.0	1022.0	1196.0	0.8545	0.8520	972.0	975.0	1267.0	0.7695	0.7672
0.0	20.0	1040	1.7562	0.0026	6270.8480	4346.6206	1998.0	2475.0	0.8073	2000.0	0.8081	1020.0	1023.0	1196.0	0.8554	0.8528	972.0	975.0	1267.0	0.7695	0.7672

Framework versions

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