GSM8K-Binary_Llama-3.2-1B-gt3yb3cm

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: 1.4064
• Model Preparation Time: 0.0055
• Mdl: 5021.6739
• Accumulated Loss: 3480.7591
• Correct Preds: 1544.0
• Total Preds: 2475.0
• Accuracy: 0.6238
• Correct Gen Preds: 261.0
• Gen Accuracy: 0.1055
• Correct Gen Preds 34192: 99.0
• Correct Preds 34192: 991.0
• Total Labels 34192: 1196.0
• Accuracy 34192: 0.8286
• Gen Accuracy 34192: 0.0828
• Correct Gen Preds 41568: 155.0
• Correct Preds 41568: 553.0
• Total Labels 41568: 1267.0
• Accuracy 41568: 0.4365
• Gen Accuracy 41568: 0.1223

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	0.0055	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
2.5224	1.0	2	2.3937	0.0055	8547.2902	5924.5301	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
1.7841	2.0	4	3.2020	0.0055	11433.4129	7925.0379	1196.0	2475.0	0.4832	8.0	0.0032	0.0	1196.0	1196.0	1.0	0.0	0.0	0.0	1267.0	0.0	0.0
0.8475	3.0	6	0.8616	0.0055	3076.3645	2132.3734	1266.0	2475.0	0.5115	7.0	0.0028	0.0	0.0	1196.0	0.0	0.0	0.0	1266.0	1267.0	0.9992	0.0
0.9462	4.0	8	0.8428	0.0055	3009.1929	2085.8135	1307.0	2475.0	0.5281	7.0	0.0028	0.0	520.0	1196.0	0.4348	0.0	0.0	787.0	1267.0	0.6212	0.0
0.9051	5.0	10	0.7595	0.0055	2711.8650	1879.7216	1330.0	2475.0	0.5374	7.0	0.0028	0.0	182.0	1196.0	0.1522	0.0	0.0	1148.0	1267.0	0.9061	0.0
1.812	6.0	12	0.8649	0.0055	3088.4505	2140.7508	1196.0	2475.0	0.4832	7.0	0.0028	0.0	1196.0	1196.0	1.0	0.0	0.0	0.0	1267.0	0.0	0.0
1.6264	7.0	14	0.7345	0.0055	2622.7585	1817.9576	1357.0	2475.0	0.5483	7.0	0.0028	0.0	858.0	1196.0	0.7174	0.0	0.0	499.0	1267.0	0.3938	0.0
0.5291	8.0	16	0.7757	0.0055	2769.7376	1919.8358	1403.0	2475.0	0.5669	7.0	0.0028	0.0	1022.0	1196.0	0.8545	0.0	0.0	381.0	1267.0	0.3007	0.0
0.237	9.0	18	0.8259	0.0055	2949.0298	2044.1117	1514.0	2475.0	0.6117	20.0	0.0081	0.0	973.0	1196.0	0.8135	0.0	12.0	541.0	1267.0	0.4270	0.0095
0.0421	10.0	20	1.3945	0.0055	4979.2689	3451.3662	1390.0	2475.0	0.5616	478.0	0.1931	1.0	367.0	1196.0	0.3069	0.0008	470.0	1023.0	1267.0	0.8074	0.3710
0.1491	11.0	22	1.4601	0.0055	5213.3876	3613.6449	1524.0	2475.0	0.6158	59.0	0.0238	13.0	1061.0	1196.0	0.8871	0.0109	39.0	463.0	1267.0	0.3654	0.0308
1.1113	12.0	24	1.4064	0.0055	5021.6739	3480.7591	1544.0	2475.0	0.6238	261.0	0.1055	99.0	991.0	1196.0	0.8286	0.0828	155.0	553.0	1267.0	0.4365	0.1223
0.0007	13.0	26	1.9100	0.0055	6819.9622	4727.2376	1436.0	2475.0	0.5802	858.0	0.3467	74.0	398.0	1196.0	0.3328	0.0619	776.0	1038.0	1267.0	0.8193	0.6125
0.0082	14.0	28	1.7320	0.0055	6184.5284	4286.7884	1502.0	2475.0	0.6069	960.0	0.3879	355.0	735.0	1196.0	0.6145	0.2968	597.0	767.0	1267.0	0.6054	0.4712
1.0707	15.0	30	2.0597	0.0055	7354.4002	5097.6817	1511.0	2475.0	0.6105	1248.0	0.5042	725.0	917.0	1196.0	0.7667	0.6062	516.0	594.0	1267.0	0.4688	0.4073
1.0696	16.0	32	2.1505	0.0055	7678.6998	5322.4691	1515.0	2475.0	0.6121	1359.0	0.5491	767.0	878.0	1196.0	0.7341	0.6413	584.0	637.0	1267.0	0.5028	0.4609
1.0693	17.0	34	2.1829	0.0055	7794.3632	5402.6409	1523.0	2475.0	0.6154	1395.0	0.5636	749.0	845.0	1196.0	0.7065	0.6263	638.0	678.0	1267.0	0.5351	0.5036
0.0	18.0	36	2.2042	0.0055	7870.3884	5455.3375	1521.0	2475.0	0.6145	1413.0	0.5709	732.0	810.0	1196.0	0.6773	0.6120	673.0	711.0	1267.0	0.5612	0.5312
0.0	19.0	38	2.2162	0.0055	7913.3941	5485.1468	1523.0	2475.0	0.6154	1425.0	0.5758	715.0	785.0	1196.0	0.6564	0.5978	703.0	738.0	1267.0	0.5825	0.5549
0.0	20.0	40	2.2249	0.0055	7944.3650	5506.6142	1513.0	2475.0	0.6113	1417.0	0.5725	697.0	763.0	1196.0	0.6380	0.5828	713.0	750.0	1267.0	0.5919	0.5627
0.0	21.0	42	2.2290	0.0055	7958.9258	5516.7070	1516.0	2475.0	0.6125	1426.0	0.5762	693.0	754.0	1196.0	0.6304	0.5794	726.0	762.0	1267.0	0.6014	0.5730
0.0	22.0	44	2.2335	0.0055	7975.2389	5528.0144	1517.0	2475.0	0.6129	1432.0	0.5786	691.0	750.0	1196.0	0.6271	0.5778	734.0	767.0	1267.0	0.6054	0.5793
0.0	23.0	46	2.2385	0.0055	7992.7922	5540.1814	1518.0	2475.0	0.6133	1435.0	0.5798	690.0	748.0	1196.0	0.6254	0.5769	738.0	770.0	1267.0	0.6077	0.5825
0.0	24.0	48	2.2436	0.0055	8011.1401	5552.8992	1515.0	2475.0	0.6121	1439.0	0.5814	694.0	748.0	1196.0	0.6254	0.5803	737.0	767.0	1267.0	0.6054	0.5817
0.0	25.0	50	2.2498	0.0055	8033.4137	5568.3380	1512.0	2475.0	0.6109	1431.0	0.5782	691.0	746.0	1196.0	0.6237	0.5778	733.0	766.0	1267.0	0.6046	0.5785
0.0	26.0	52	2.2514	0.0055	8039.0644	5572.2548	1520.0	2475.0	0.6141	1443.0	0.5830	699.0	753.0	1196.0	0.6296	0.5844	736.0	767.0	1267.0	0.6054	0.5809
0.0	27.0	54	2.2563	0.0055	8056.4092	5584.2773	1516.0	2475.0	0.6125	1438.0	0.5810	697.0	751.0	1196.0	0.6279	0.5828	734.0	765.0	1267.0	0.6038	0.5793
0.0	28.0	56	2.2602	0.0055	8070.5630	5594.0880	1517.0	2475.0	0.6129	1443.0	0.5830	701.0	753.0	1196.0	0.6296	0.5861	734.0	764.0	1267.0	0.6030	0.5793
0.0	29.0	58	2.2622	0.0055	8077.6162	5598.9769	1513.0	2475.0	0.6113	1433.0	0.5790	697.0	752.0	1196.0	0.6288	0.5828	729.0	761.0	1267.0	0.6006	0.5754
0.0	30.0	60	2.2649	0.0055	8087.1150	5605.5610	1518.0	2475.0	0.6133	1440.0	0.5818	704.0	757.0	1196.0	0.6329	0.5886	729.0	761.0	1267.0	0.6006	0.5754
0.0	31.0	62	2.2656	0.0055	8089.6775	5607.3372	1519.0	2475.0	0.6137	1441.0	0.5822	704.0	758.0	1196.0	0.6338	0.5886	730.0	761.0	1267.0	0.6006	0.5762
0.0	32.0	64	2.2689	0.0055	8101.4802	5615.5182	1514.0	2475.0	0.6117	1439.0	0.5814	703.0	756.0	1196.0	0.6321	0.5878	728.0	758.0	1267.0	0.5983	0.5746
0.0	33.0	66	2.2710	0.0055	8108.8518	5620.6278	1516.0	2475.0	0.6125	1436.0	0.5802	702.0	758.0	1196.0	0.6338	0.5870	727.0	758.0	1267.0	0.5983	0.5738
0.0	34.0	68	2.2741	0.0055	8119.9634	5628.3297	1518.0	2475.0	0.6133	1438.0	0.5810	704.0	760.0	1196.0	0.6355	0.5886	727.0	758.0	1267.0	0.5983	0.5738
1.0693	35.0	70	2.2735	0.0055	8117.9366	5626.9249	1517.0	2475.0	0.6129	1437.0	0.5806	705.0	760.0	1196.0	0.6355	0.5895	725.0	757.0	1267.0	0.5975	0.5722
0.0	36.0	72	2.2753	0.0055	8124.4509	5631.4402	1516.0	2475.0	0.6125	1436.0	0.5802	704.0	762.0	1196.0	0.6371	0.5886	725.0	754.0	1267.0	0.5951	0.5722
0.0	37.0	74	2.2753	0.0055	8124.3745	5631.3873	1514.0	2475.0	0.6117	1436.0	0.5802	707.0	762.0	1196.0	0.6371	0.5911	721.0	752.0	1267.0	0.5935	0.5691
0.0	38.0	76	2.2788	0.0055	8136.6842	5639.9197	1521.0	2475.0	0.6145	1441.0	0.5822	710.0	766.0	1196.0	0.6405	0.5936	724.0	755.0	1267.0	0.5959	0.5714
0.0	39.0	78	2.2790	0.0055	8137.4584	5640.4564	1516.0	2475.0	0.6125	1438.0	0.5810	710.0	764.0	1196.0	0.6388	0.5936	720.0	752.0	1267.0	0.5935	0.5683
0.0	40.0	80	2.2802	0.0055	8141.8020	5643.4671	1517.0	2475.0	0.6129	1438.0	0.5810	710.0	766.0	1196.0	0.6405	0.5936	721.0	751.0	1267.0	0.5927	0.5691
0.0	41.0	82	2.2808	0.0055	8144.0353	5645.0151	1519.0	2475.0	0.6137	1439.0	0.5814	711.0	767.0	1196.0	0.6413	0.5945	720.0	752.0	1267.0	0.5935	0.5683
0.0	42.0	84	2.2831	0.0055	8152.0846	5650.5945	1516.0	2475.0	0.6125	1439.0	0.5814	710.0	765.0	1196.0	0.6396	0.5936	721.0	751.0	1267.0	0.5927	0.5691

Framework versions

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