logical-reasoning / data /Qwen2.5-72B-Instruct_metrics.csv
dh-mc's picture
ready for final run
8157c36
raw
history blame
1.81 kB
epoch,model,run,accuracy,precision,recall,f1,ratio_valid_classifications
0.0,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct_torch.bfloat16_4bit_lf,0.7956666666666666,0.8098073411161181,0.7956666666666666,0.7771317592221199,0.994
0.2,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-35_torch.bfloat16_4bit_lf,0.792,0.8180793658647517,0.792,0.80166512366027,1.0
0.4,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-70_torch.bfloat16_4bit_lf,0.7716666666666666,0.8199569804721152,0.7716666666666666,0.7895879011938259,1.0
0.6,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-105_torch.bfloat16_4bit_lf,0.798,0.8379062379534957,0.798,0.812148680520218,1.0
0.8,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-140_torch.bfloat16_4bit_lf,0.8213333333333334,0.8447926258362122,0.8213333333333334,0.8299486611547571,1.0
1.0,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-175_torch.bfloat16_4bit_lf,0.7643333333333333,0.8235366724638146,0.7643333333333333,0.7858148913986999,1.0
1.2,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-210_torch.bfloat16_4bit_lf,0.7986666666666666,0.83233218480008,0.7986666666666666,0.8115886421806521,1.0
1.4,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-245_torch.bfloat16_4bit_lf,0.7923333333333333,0.8231874218285514,0.7923333333333333,0.803363661387202,1.0
1.6,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-280_torch.bfloat16_4bit_lf,0.7936666666666666,0.8268750473800219,0.7936666666666666,0.8057720333101867,1.0
1.8,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-315_torch.bfloat16_4bit_lf,0.801,0.830389411421043,0.801,0.8117656427717702,1.0
2.0,Qwen2.5-72B-Instruct,Qwen/Qwen2.5-72B-Instruct/checkpoint-350_torch.bfloat16_4bit_lf,0.795,0.8280696193638868,0.795,0.8068114730639832,1.0