GSM8K-Binary_Llama-3.2-1B-n45gfm9o

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: 1.4701
• Model Preparation Time: 0.0059
• Mdl: 5249.0917
• Accumulated Loss: 3638.3931
• Correct Preds: 1954.0
• Total Preds: 2475.0
• Accuracy: 0.7895
• Correct Gen Preds: 1866.0
• Gen Accuracy: 0.7539
• Correct Gen Preds 34192: 949.0
• Correct Preds 34192: 1014.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8478
• Gen Accuracy 34192: 0.7935
• Correct Gen Preds 41568: 908.0
• Correct Preds 41568: 940.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.7419
• Gen Accuracy 41568: 0.7167

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
0.774	1.0	36	0.7619	0.0059	2720.3405	1885.5964	1654.0	2475.0	0.6683	8.0	0.0032	0.0	565.0	1196.0	0.4724	0.0	0.0	1089.0	1267.0	0.8595	0.0
0.558	2.0	72	0.7026	0.0059	2508.8193	1738.9811	1704.0	2475.0	0.6885	8.0	0.0032	0.0	527.0	1196.0	0.4406	0.0	0.0	1177.0	1267.0	0.9290	0.0
0.3275	3.0	108	0.7695	0.0059	2747.8047	1904.6331	1687.0	2475.0	0.6816	16.0	0.0065	0.0	533.0	1196.0	0.4457	0.0	8.0	1154.0	1267.0	0.9108	0.0063
0.2039	4.0	144	0.6071	0.0059	2167.6033	1502.4681	1932.0	2475.0	0.7806	71.0	0.0287	2.0	916.0	1196.0	0.7659	0.0017	61.0	1016.0	1267.0	0.8019	0.0481
0.2072	5.0	180	1.0169	0.0059	3630.9726	2516.7984	1924.0	2475.0	0.7774	1267.0	0.5119	597.0	1044.0	1196.0	0.8729	0.4992	661.0	880.0	1267.0	0.6946	0.5217
0.066	6.0	216	1.0546	0.0059	3765.5184	2610.0584	1893.0	2475.0	0.7648	1017.0	0.4109	500.0	1079.0	1196.0	0.9022	0.4181	509.0	814.0	1267.0	0.6425	0.4017
0.0004	7.0	252	1.2220	0.0059	4363.2817	3024.3964	1906.0	2475.0	0.7701	1454.0	0.5875	544.0	850.0	1196.0	0.7107	0.4548	902.0	1056.0	1267.0	0.8335	0.7119
0.0021	8.0	288	1.7093	0.0059	6103.2758	4230.4684	1892.0	2475.0	0.7644	1727.0	0.6978	1018.0	1103.0	1196.0	0.9222	0.8512	701.0	789.0	1267.0	0.6227	0.5533
0.0002	9.0	324	1.5981	0.0059	5706.2934	3955.3012	1900.0	2475.0	0.7677	1761.0	0.7115	1001.0	1078.0	1196.0	0.9013	0.8370	752.0	822.0	1267.0	0.6488	0.5935
0.0001	10.0	360	1.4701	0.0059	5249.0917	3638.3931	1954.0	2475.0	0.7895	1866.0	0.7539	949.0	1014.0	1196.0	0.8478	0.7935	908.0	940.0	1267.0	0.7419	0.7167
0.0	11.0	396	1.4879	0.0059	5312.7263	3682.5013	1950.0	2475.0	0.7879	1869.0	0.7552	958.0	1020.0	1196.0	0.8528	0.8010	903.0	930.0	1267.0	0.7340	0.7127
0.0	12.0	432	1.4948	0.0059	5337.3050	3699.5379	1948.0	2475.0	0.7871	1867.0	0.7543	960.0	1022.0	1196.0	0.8545	0.8027	898.0	926.0	1267.0	0.7309	0.7088
0.0	13.0	468	1.5004	0.0059	5357.5988	3713.6045	1946.0	2475.0	0.7863	1866.0	0.7539	961.0	1024.0	1196.0	0.8562	0.8035	896.0	922.0	1267.0	0.7277	0.7072
0.7841	14.0	504	1.5063	0.0059	5378.5450	3728.1233	1948.0	2475.0	0.7871	1871.0	0.7560	966.0	1026.0	1196.0	0.8579	0.8077	896.0	922.0	1267.0	0.7277	0.7072
0.0	15.0	540	1.5092	0.0059	5388.9284	3735.3205	1945.0	2475.0	0.7859	1871.0	0.7560	970.0	1026.0	1196.0	0.8579	0.8110	893.0	919.0	1267.0	0.7253	0.7048
0.7841	16.0	576	1.5127	0.0059	5401.2640	3743.8709	1944.0	2475.0	0.7855	1869.0	0.7552	968.0	1025.0	1196.0	0.8570	0.8094	892.0	919.0	1267.0	0.7253	0.7040
0.0	17.0	612	1.5144	0.0059	5407.4933	3748.1887	1943.0	2475.0	0.7851	1875.0	0.7576	973.0	1026.0	1196.0	0.8579	0.8135	893.0	917.0	1267.0	0.7238	0.7048
0.0	18.0	648	1.5177	0.0059	5419.3768	3756.4257	1947.0	2475.0	0.7867	1872.0	0.7564	972.0	1029.0	1196.0	0.8604	0.8127	892.0	918.0	1267.0	0.7245	0.7040
0.0	19.0	684	1.5195	0.0059	5425.7845	3760.8672	1945.0	2475.0	0.7859	1877.0	0.7584	976.0	1028.0	1196.0	0.8595	0.8161	893.0	917.0	1267.0	0.7238	0.7048
0.0	20.0	720	1.5238	0.0059	5441.1232	3771.4992	1944.0	2475.0	0.7855	1875.0	0.7576	972.0	1027.0	1196.0	0.8587	0.8127	895.0	917.0	1267.0	0.7238	0.7064
0.0	21.0	756	1.5265	0.0059	5450.7090	3778.1436	1948.0	2475.0	0.7871	1881.0	0.76	979.0	1031.0	1196.0	0.8620	0.8186	893.0	917.0	1267.0	0.7238	0.7048
0.0	22.0	792	1.5266	0.0059	5450.9628	3778.3195	1944.0	2475.0	0.7855	1874.0	0.7572	974.0	1029.0	1196.0	0.8604	0.8144	891.0	915.0	1267.0	0.7222	0.7032
0.0	23.0	828	1.5306	0.0059	5465.2671	3788.2345	1947.0	2475.0	0.7867	1877.0	0.7584	979.0	1032.0	1196.0	0.8629	0.8186	890.0	915.0	1267.0	0.7222	0.7024
0.0	24.0	864	1.5319	0.0059	5469.8094	3791.3829	1950.0	2475.0	0.7879	1882.0	0.7604	981.0	1033.0	1196.0	0.8637	0.8202	893.0	917.0	1267.0	0.7238	0.7048

Framework versions

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