ARC-Easy_Llama-3.2-1B-1e9mspm5

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: 0.6837
• Model Preparation Time: 0.0059
• Mdl: 562.2638
• Accumulated Loss: 389.7316
• Correct Preds: 444.0
• Total Preds: 570.0
• Accuracy: 0.7789
• Correct Gen Preds: 0.0
• Gen Accuracy: 0.0
• Correct Gen Preds 32: 0.0
• Correct Preds 32: 119.0
• Total Labels 32: 158.0
• Accuracy 32: 0.7532
• Gen Accuracy 32: 0.0
• Correct Gen Preds 33: 0.0
• Correct Preds 33: 126.0
• Total Labels 33: 152.0
• Accuracy 33: 0.8289
• Gen Accuracy 33: 0.0
• Correct Gen Preds 34: 0.0
• Correct Preds 34: 111.0
• Total Labels 34: 142.0
• Accuracy 34: 0.7817
• Gen Accuracy 34: 0.0
• Correct Gen Preds 35: 0.0
• Correct Preds 35: 88.0
• Total Labels 35: 118.0
• Accuracy 35: 0.7458
• Gen Accuracy 35: 0.0
• Correct Gen Preds 36: 0.0
• Correct Preds 36: 0.0
• Total Labels 36: 0.0
• Accuracy 36: 0.0
• Gen Accuracy 36: 0.0

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: 64
• eval_batch_size: 112
• 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 32	Correct Preds 32	Total Labels 32	Accuracy 32	Gen Accuracy 32	Correct Gen Preds 33	Correct Preds 33	Total Labels 33	Accuracy 33	Gen Accuracy 33	Correct Gen Preds 34	Correct Preds 34	Total Labels 34	Accuracy 34	Gen Accuracy 34	Correct Gen Preds 35	Correct Preds 35	Total Labels 35	Accuracy 35	Gen Accuracy 35	Correct Gen Preds 36	Correct Preds 36	Total Labels 36	Accuracy 36	Gen Accuracy 36
No log	0	0	1.5354	0.0059	1262.6022	875.1692	172.0	570.0	0.3018	170.0	0.2982	154.0	154.0	158.0	0.9747	0.9747	0.0	0.0	152.0	0.0	0.0	15.0	17.0	142.0	0.1197	0.1056	1.0	1.0	118.0	0.0085	0.0085	0.0	0.0	0.0	0.0	0.0
0.4429	1.0	36	0.6869	0.0059	564.8439	391.5200	430.0	570.0	0.7544	60.0	0.1053	0.0	130.0	158.0	0.8228	0.0	10.0	113.0	152.0	0.7434	0.0658	43.0	108.0	142.0	0.7606	0.3028	7.0	79.0	118.0	0.6695	0.0593	0.0	0.0	0.0	0.0	0.0
0.1294	2.0	72	0.6837	0.0059	562.2638	389.7316	444.0	570.0	0.7789	0.0	0.0	0.0	119.0	158.0	0.7532	0.0	0.0	126.0	152.0	0.8289	0.0	0.0	111.0	142.0	0.7817	0.0	0.0	88.0	118.0	0.7458	0.0	0.0	0.0	0.0	0.0	0.0
0.0423	3.0	108	0.9005	0.0059	740.5548	513.3135	418.0	570.0	0.7333	411.0	0.7211	110.0	116.0	158.0	0.7342	0.6962	128.0	128.0	152.0	0.8421	0.8421	102.0	103.0	142.0	0.7254	0.7183	71.0	71.0	118.0	0.6017	0.6017	0.0	0.0	0.0	0.0	0.0
0.0101	4.0	144	1.1974	0.0059	984.6596	682.5140	429.0	570.0	0.7526	111.0	0.1947	0.0	117.0	158.0	0.7405	0.0	49.0	124.0	152.0	0.8158	0.3224	59.0	109.0	142.0	0.7676	0.4155	3.0	79.0	118.0	0.6695	0.0254	0.0	0.0	0.0	0.0	0.0
0.0022	5.0	180	1.9793	0.0059	1627.6320	1128.1885	428.0	570.0	0.7509	384.0	0.6737	85.0	112.0	158.0	0.7089	0.5380	118.0	119.0	152.0	0.7829	0.7763	109.0	109.0	142.0	0.7676	0.7676	72.0	88.0	118.0	0.7458	0.6102	0.0	0.0	0.0	0.0	0.0
0.0004	6.0	216	2.1635	0.0059	1779.1008	1233.1787	440.0	570.0	0.7719	236.0	0.4140	17.0	126.0	158.0	0.7975	0.1076	79.0	118.0	152.0	0.7763	0.5197	92.0	112.0	142.0	0.7887	0.6479	48.0	84.0	118.0	0.7119	0.4068	0.0	0.0	0.0	0.0	0.0

Framework versions

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