| [ | |
| { | |
| "loss": 0.6629, | |
| "grad_norm": 1.981173038482666, | |
| "learning_rate": 1.8575498575498575e-05, | |
| "epoch": 0.7122507122507122, | |
| "step": 500 | |
| }, | |
| { | |
| "eval_loss": 0.3868238925933838, | |
| "eval_accuracy": 0.8307254623044097, | |
| "eval_runtime": 3.5243, | |
| "eval_samples_per_second": 199.474, | |
| "eval_steps_per_second": 49.939, | |
| "epoch": 1.0, | |
| "step": 702 | |
| }, | |
| { | |
| "loss": 0.4271, | |
| "grad_norm": 0.7268210053443909, | |
| "learning_rate": 1.7150997150997152e-05, | |
| "epoch": 1.4245014245014245, | |
| "step": 1000 | |
| }, | |
| { | |
| "eval_loss": 0.12105754762887955, | |
| "eval_accuracy": 0.9544807965860598, | |
| "eval_runtime": 4.3029, | |
| "eval_samples_per_second": 163.38, | |
| "eval_steps_per_second": 40.903, | |
| "epoch": 2.0, | |
| "step": 1404 | |
| }, | |
| { | |
| "loss": 0.1895, | |
| "grad_norm": 0.8351957201957703, | |
| "learning_rate": 1.5726495726495726e-05, | |
| "epoch": 2.1367521367521367, | |
| "step": 1500 | |
| }, | |
| { | |
| "loss": 0.1534, | |
| "grad_norm": 0.24414250254631042, | |
| "learning_rate": 1.4301994301994305e-05, | |
| "epoch": 2.849002849002849, | |
| "step": 2000 | |
| }, | |
| { | |
| "eval_loss": 0.09802740812301636, | |
| "eval_accuracy": 0.9701280227596017, | |
| "eval_runtime": 3.5319, | |
| "eval_samples_per_second": 199.043, | |
| "eval_steps_per_second": 49.832, | |
| "epoch": 3.0, | |
| "step": 2106 | |
| }, | |
| { | |
| "loss": 0.1275, | |
| "grad_norm": 0.11179369688034058, | |
| "learning_rate": 1.2877492877492879e-05, | |
| "epoch": 3.561253561253561, | |
| "step": 2500 | |
| }, | |
| { | |
| "eval_loss": 0.093312107026577, | |
| "eval_accuracy": 0.9772403982930299, | |
| "eval_runtime": 3.7294, | |
| "eval_samples_per_second": 188.501, | |
| "eval_steps_per_second": 47.192, | |
| "epoch": 4.0, | |
| "step": 2808 | |
| }, | |
| { | |
| "loss": 0.1302, | |
| "grad_norm": 0.037690628319978714, | |
| "learning_rate": 1.1452991452991454e-05, | |
| "epoch": 4.273504273504273, | |
| "step": 3000 | |
| }, | |
| { | |
| "loss": 0.1332, | |
| "grad_norm": 0.06977783888578415, | |
| "learning_rate": 1.002849002849003e-05, | |
| "epoch": 4.985754985754986, | |
| "step": 3500 | |
| }, | |
| { | |
| "eval_loss": 0.09014463424682617, | |
| "eval_accuracy": 0.9786628733997155, | |
| "eval_runtime": 3.6975, | |
| "eval_samples_per_second": 190.126, | |
| "eval_steps_per_second": 47.599, | |
| "epoch": 5.0, | |
| "step": 3510 | |
| }, | |
| { | |
| "loss": 0.1051, | |
| "grad_norm": 0.07191012799739838, | |
| "learning_rate": 8.603988603988605e-06, | |
| "epoch": 5.698005698005698, | |
| "step": 4000 | |
| }, | |
| { | |
| "eval_loss": 0.08116021007299423, | |
| "eval_accuracy": 0.9800853485064012, | |
| "eval_runtime": 3.558, | |
| "eval_samples_per_second": 197.582, | |
| "eval_steps_per_second": 49.466, | |
| "epoch": 6.0, | |
| "step": 4212 | |
| }, | |
| { | |
| "loss": 0.1026, | |
| "grad_norm": 9.66889476776123, | |
| "learning_rate": 7.17948717948718e-06, | |
| "epoch": 6.410256410256411, | |
| "step": 4500 | |
| }, | |
| { | |
| "eval_loss": 0.07601884752511978, | |
| "eval_accuracy": 0.9800853485064012, | |
| "eval_runtime": 7.4677, | |
| "eval_samples_per_second": 94.138, | |
| "eval_steps_per_second": 23.568, | |
| "epoch": 7.0, | |
| "step": 4914 | |
| }, | |
| { | |
| "loss": 0.092, | |
| "grad_norm": 0.02262728288769722, | |
| "learning_rate": 5.7549857549857555e-06, | |
| "epoch": 7.122507122507122, | |
| "step": 5000 | |
| }, | |
| { | |
| "loss": 0.0864, | |
| "grad_norm": 0.1124846562743187, | |
| "learning_rate": 4.330484330484331e-06, | |
| "epoch": 7.834757834757835, | |
| "step": 5500 | |
| }, | |
| { | |
| "eval_loss": 0.07228563725948334, | |
| "eval_accuracy": 0.9800853485064012, | |
| "eval_runtime": 4.228, | |
| "eval_samples_per_second": 166.273, | |
| "eval_steps_per_second": 41.627, | |
| "epoch": 8.0, | |
| "step": 5616 | |
| }, | |
| { | |
| "loss": 0.1273, | |
| "grad_norm": 0.0037623795215040445, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "epoch": 8.547008547008547, | |
| "step": 6000 | |
| }, | |
| { | |
| "eval_loss": 0.0683717355132103, | |
| "eval_accuracy": 0.9815078236130867, | |
| "eval_runtime": 3.5474, | |
| "eval_samples_per_second": 198.175, | |
| "eval_steps_per_second": 49.614, | |
| "epoch": 9.0, | |
| "step": 6318 | |
| }, | |
| { | |
| "loss": 0.0724, | |
| "grad_norm": 0.004779215436428785, | |
| "learning_rate": 1.4814814814814815e-06, | |
| "epoch": 9.25925925925926, | |
| "step": 6500 | |
| }, | |
| { | |
| "loss": 0.1116, | |
| "grad_norm": 2.0302906036376953, | |
| "learning_rate": 5.6980056980056986e-08, | |
| "epoch": 9.971509971509972, | |
| "step": 7000 | |
| }, | |
| { | |
| "eval_loss": 0.06746786087751389, | |
| "eval_accuracy": 0.9815078236130867, | |
| "eval_runtime": 7.6005, | |
| "eval_samples_per_second": 92.494, | |
| "eval_steps_per_second": 23.156, | |
| "epoch": 10.0, | |
| "step": 7020 | |
| }, | |
| { | |
| "train_runtime": 449.8629, | |
| "train_samples_per_second": 62.419, | |
| "train_steps_per_second": 15.605, | |
| "total_flos": 520823902075200.0, | |
| "train_loss": 0.17984749846308998, | |
| "epoch": 10.0, | |
| "step": 7020 | |
| } | |
| ] |