GSM8K-Binary_Llama-3.2-1B-trn9haqb

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: 1.3861
• Model Preparation Time: 0.0059
• Mdl: 4949.1510
• Accumulated Loss: 3430.4901
• Correct Preds: 1952.0
• Total Preds: 2475.0
• Accuracy: 0.7887
• Correct Gen Preds: 1959.0
• Gen Accuracy: 0.7915
• Correct Gen Preds 34192: 979.0
• Correct Preds 34192: 980.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8194
• Gen Accuracy 34192: 0.8186
• Correct Gen Preds 41568: 971.0
• Correct Preds 41568: 972.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.7672
• Gen Accuracy 41568: 0.7664

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: constant
• lr_scheduler_warmup_ratio: 0.001
• 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	0.0059	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
0.5562	1.0	33	0.5791	0.0059	2067.7129	1433.2294	1800.0	2475.0	0.7273	136.0	0.0549	0.0	1001.0	1196.0	0.8370	0.0	129.0	799.0	1267.0	0.6306	0.1018
0.2241	2.0	66	0.6917	0.0059	2469.9161	1712.0154	1750.0	2475.0	0.7071	25.0	0.0101	2.0	1150.0	1196.0	0.9615	0.0017	16.0	600.0	1267.0	0.4736	0.0126
0.2445	3.0	99	0.5807	0.0059	2073.3593	1437.1432	1901.0	2475.0	0.7681	672.0	0.2715	56.0	817.0	1196.0	0.6831	0.0468	608.0	1084.0	1267.0	0.8556	0.4799
0.3792	4.0	132	1.0107	0.0059	3608.9732	2501.5496	1808.0	2475.0	0.7305	752.0	0.3038	361.0	1124.0	1196.0	0.9398	0.3018	383.0	684.0	1267.0	0.5399	0.3023
0.5311	5.0	165	1.0453	0.0059	3732.4069	2587.1073	1949.0	2475.0	0.7875	1925.0	0.7778	965.0	987.0	1196.0	0.8253	0.8069	951.0	962.0	1267.0	0.7593	0.7506
0.0003	6.0	198	1.1808	0.0059	4216.3306	2922.5376	1929.0	2475.0	0.7794	1889.0	0.7632	997.0	1034.0	1196.0	0.8645	0.8336	885.0	895.0	1267.0	0.7064	0.6985
0.4714	7.0	231	1.7950	0.0059	6409.2129	4442.5278	1910.0	2475.0	0.7717	1900.0	0.7677	1077.0	1092.0	1196.0	0.9130	0.9005	815.0	818.0	1267.0	0.6456	0.6433
0.0002	8.0	264	1.3861	0.0059	4949.1510	3430.4901	1952.0	2475.0	0.7887	1959.0	0.7915	979.0	980.0	1196.0	0.8194	0.8186	971.0	972.0	1267.0	0.7672	0.7664
0.0001	9.0	297	1.8078	0.0059	6455.1510	4474.3697	1889.0	2475.0	0.7632	1895.0	0.7657	1088.0	1089.0	1196.0	0.9105	0.9097	799.0	800.0	1267.0	0.6314	0.6306
0.0	10.0	330	1.6442	0.0059	5870.8161	4069.3396	1937.0	2475.0	0.7826	1944.0	0.7855	1059.0	1059.0	1196.0	0.8855	0.8855	877.0	878.0	1267.0	0.6930	0.6922
0.0	11.0	363	1.6431	0.0059	5866.8306	4066.5771	1938.0	2475.0	0.7830	1946.0	0.7863	1058.0	1058.0	1196.0	0.8846	0.8846	880.0	880.0	1267.0	0.6946	0.6946
0.0	12.0	396	1.6410	0.0059	5859.5168	4061.5076	1934.0	2475.0	0.7814	1941.0	0.7842	1055.0	1055.0	1196.0	0.8821	0.8821	878.0	879.0	1267.0	0.6938	0.6930
0.0	13.0	429	1.6420	0.0059	5863.0062	4063.9262	1935.0	2475.0	0.7818	1943.0	0.7851	1056.0	1056.0	1196.0	0.8829	0.8829	879.0	879.0	1267.0	0.6938	0.6938
0.0	14.0	462	1.6393	0.0059	5853.5075	4057.3422	1936.0	2475.0	0.7822	1944.0	0.7855	1055.0	1055.0	1196.0	0.8821	0.8821	881.0	881.0	1267.0	0.6953	0.6953
0.4705	15.0	495	1.6394	0.0059	5853.9322	4057.6366	1935.0	2475.0	0.7818	1943.0	0.7851	1054.0	1054.0	1196.0	0.8813	0.8813	881.0	881.0	1267.0	0.6953	0.6953
0.0	16.0	528	1.6388	0.0059	5851.4802	4055.9370	1936.0	2475.0	0.7822	1944.0	0.7855	1055.0	1055.0	1196.0	0.8821	0.8821	881.0	881.0	1267.0	0.6953	0.6953
0.0	17.0	561	1.6396	0.0059	5854.5643	4058.0747	1937.0	2475.0	0.7826	1945.0	0.7859	1054.0	1054.0	1196.0	0.8813	0.8813	883.0	883.0	1267.0	0.6969	0.6969
0.4705	18.0	594	1.6388	0.0059	5851.7692	4056.1373	1937.0	2475.0	0.7826	1945.0	0.7859	1053.0	1053.0	1196.0	0.8804	0.8804	884.0	884.0	1267.0	0.6977	0.6977
0.0	19.0	627	1.6396	0.0059	5854.6347	4058.1235	1935.0	2475.0	0.7818	1943.0	0.7851	1052.0	1052.0	1196.0	0.8796	0.8796	883.0	883.0	1267.0	0.6969	0.6969
0.0	20.0	660	1.6372	0.0059	5845.9689	4052.1169	1936.0	2475.0	0.7822	1944.0	0.7855	1052.0	1052.0	1196.0	0.8796	0.8796	884.0	884.0	1267.0	0.6977	0.6977
0.0	21.0	693	1.6389	0.0059	5852.0283	4056.3169	1935.0	2475.0	0.7818	1943.0	0.7851	1052.0	1052.0	1196.0	0.8796	0.8796	883.0	883.0	1267.0	0.6969	0.6969
0.0	22.0	726	1.6393	0.0059	5853.4144	4057.2777	1936.0	2475.0	0.7822	1944.0	0.7855	1053.0	1053.0	1196.0	0.8804	0.8804	883.0	883.0	1267.0	0.6969	0.6969
0.0	23.0	759	1.6391	0.0059	5852.5099	4056.6507	1934.0	2475.0	0.7814	1942.0	0.7846	1051.0	1051.0	1196.0	0.8788	0.8788	883.0	883.0	1267.0	0.6969	0.6969

Framework versions

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