| { | |
| "best_metric": 0.9595276897831222, | |
| "best_model_checkpoint": "dino-vitb8-finetuned-stroke-binary/checkpoint-2900", | |
| "epoch": 17.905426356589146, | |
| "eval_steps": 100, | |
| "global_step": 2900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31007751937984496, | |
| "grad_norm": 6.24950647354126, | |
| "learning_rate": 1.6896551724137933e-06, | |
| "loss": 0.8366, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "grad_norm": 5.115139007568359, | |
| "learning_rate": 3.4137931034482764e-06, | |
| "loss": 0.7965, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "eval_accuracy": 0.5382180009045681, | |
| "eval_f1": 0.5057986540639501, | |
| "eval_loss": 0.8312332630157471, | |
| "eval_precision": 0.496820169669259, | |
| "eval_recall": 0.5382180009045681, | |
| "eval_runtime": 32.0161, | |
| "eval_samples_per_second": 69.059, | |
| "eval_steps_per_second": 8.652, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 8.835868835449219, | |
| "learning_rate": 5.137931034482759e-06, | |
| "loss": 0.7713, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.235658914728682, | |
| "grad_norm": 9.565229415893555, | |
| "learning_rate": 6.862068965517242e-06, | |
| "loss": 0.6839, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.235658914728682, | |
| "eval_accuracy": 0.6246042514699232, | |
| "eval_f1": 0.5749818840100107, | |
| "eval_loss": 0.6796302199363708, | |
| "eval_precision": 0.5990946901610464, | |
| "eval_recall": 0.6246042514699232, | |
| "eval_runtime": 32.1198, | |
| "eval_samples_per_second": 68.836, | |
| "eval_steps_per_second": 8.624, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5457364341085271, | |
| "grad_norm": 8.520139694213867, | |
| "learning_rate": 8.586206896551726e-06, | |
| "loss": 0.6219, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8558139534883722, | |
| "grad_norm": 6.877398490905762, | |
| "learning_rate": 1.0310344827586208e-05, | |
| "loss": 0.5662, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.8558139534883722, | |
| "eval_accuracy": 0.7317955676164631, | |
| "eval_f1": 0.7119348319857804, | |
| "eval_loss": 0.5344149470329285, | |
| "eval_precision": 0.7376504962044594, | |
| "eval_recall": 0.7317955676164631, | |
| "eval_runtime": 32.0121, | |
| "eval_samples_per_second": 69.068, | |
| "eval_steps_per_second": 8.653, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.1612403100775195, | |
| "grad_norm": 6.749499797821045, | |
| "learning_rate": 1.203448275862069e-05, | |
| "loss": 0.4908, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.471317829457364, | |
| "grad_norm": 11.308539390563965, | |
| "learning_rate": 1.3758620689655173e-05, | |
| "loss": 0.4408, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.471317829457364, | |
| "eval_accuracy": 0.8123021257349615, | |
| "eval_f1": 0.8082421593111531, | |
| "eval_loss": 0.40815532207489014, | |
| "eval_precision": 0.8120443114553312, | |
| "eval_recall": 0.8123021257349615, | |
| "eval_runtime": 31.919, | |
| "eval_samples_per_second": 69.269, | |
| "eval_steps_per_second": 8.678, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.781395348837209, | |
| "grad_norm": 3.188601493835449, | |
| "learning_rate": 1.548275862068966e-05, | |
| "loss": 0.3761, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0868217054263565, | |
| "grad_norm": 9.05437183380127, | |
| "learning_rate": 1.7206896551724138e-05, | |
| "loss": 0.3611, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0868217054263565, | |
| "eval_accuracy": 0.8602442333785617, | |
| "eval_f1": 0.859700206427454, | |
| "eval_loss": 0.333509624004364, | |
| "eval_precision": 0.8595584495556567, | |
| "eval_recall": 0.8602442333785617, | |
| "eval_runtime": 32.0233, | |
| "eval_samples_per_second": 69.044, | |
| "eval_steps_per_second": 8.65, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3968992248062015, | |
| "grad_norm": 2.100419759750366, | |
| "learning_rate": 1.893103448275862e-05, | |
| "loss": 0.3072, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.7069767441860466, | |
| "grad_norm": 6.307981014251709, | |
| "learning_rate": 1.9999345217525786e-05, | |
| "loss": 0.3121, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.7069767441860466, | |
| "eval_accuracy": 0.8860244233378561, | |
| "eval_f1": 0.8831883391472106, | |
| "eval_loss": 0.27464836835861206, | |
| "eval_precision": 0.8913973600585926, | |
| "eval_recall": 0.8860244233378561, | |
| "eval_runtime": 31.9786, | |
| "eval_samples_per_second": 69.14, | |
| "eval_steps_per_second": 8.662, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.0124031007751935, | |
| "grad_norm": 4.161785125732422, | |
| "learning_rate": 1.999136563788985e-05, | |
| "loss": 0.2786, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.322480620155039, | |
| "grad_norm": 5.515849590301514, | |
| "learning_rate": 1.997432546533359e-05, | |
| "loss": 0.2614, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.322480620155039, | |
| "eval_accuracy": 0.9050203527815468, | |
| "eval_f1": 0.9039775360154476, | |
| "eval_loss": 0.2299066036939621, | |
| "eval_precision": 0.90576736213102, | |
| "eval_recall": 0.9050203527815468, | |
| "eval_runtime": 31.9694, | |
| "eval_samples_per_second": 69.16, | |
| "eval_steps_per_second": 8.665, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.632558139534884, | |
| "grad_norm": 2.6630172729492188, | |
| "learning_rate": 1.9948240152606154e-05, | |
| "loss": 0.2622, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.942635658914728, | |
| "grad_norm": 4.852052211761475, | |
| "learning_rate": 1.9913133354972336e-05, | |
| "loss": 0.242, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.942635658914728, | |
| "eval_accuracy": 0.9176843057440073, | |
| "eval_f1": 0.9177728397549259, | |
| "eval_loss": 0.21027454733848572, | |
| "eval_precision": 0.9179004728816644, | |
| "eval_recall": 0.9176843057440073, | |
| "eval_runtime": 31.9231, | |
| "eval_samples_per_second": 69.26, | |
| "eval_steps_per_second": 8.677, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.248062015503876, | |
| "grad_norm": 5.338088035583496, | |
| "learning_rate": 1.9869036908760982e-05, | |
| "loss": 0.2332, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.558139534883721, | |
| "grad_norm": 1.9332038164138794, | |
| "learning_rate": 1.981599080249447e-05, | |
| "loss": 0.2239, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.558139534883721, | |
| "eval_accuracy": 0.9081863410221619, | |
| "eval_f1": 0.908987571769516, | |
| "eval_loss": 0.22980067133903503, | |
| "eval_precision": 0.9136294473473457, | |
| "eval_recall": 0.9081863410221619, | |
| "eval_runtime": 31.9612, | |
| "eval_samples_per_second": 69.178, | |
| "eval_steps_per_second": 8.667, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.868217054263566, | |
| "grad_norm": 4.6057281494140625, | |
| "learning_rate": 1.9754043140625415e-05, | |
| "loss": 0.2216, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.173643410852713, | |
| "grad_norm": 3.815474271774292, | |
| "learning_rate": 1.9683250099913514e-05, | |
| "loss": 0.1979, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.173643410852713, | |
| "eval_accuracy": 0.9208502939846224, | |
| "eval_f1": 0.9197449979522355, | |
| "eval_loss": 0.20585399866104126, | |
| "eval_precision": 0.9230342997728979, | |
| "eval_recall": 0.9208502939846224, | |
| "eval_runtime": 32.0879, | |
| "eval_samples_per_second": 68.904, | |
| "eval_steps_per_second": 8.633, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.4837209302325585, | |
| "grad_norm": 3.7233753204345703, | |
| "learning_rate": 1.960367587848211e-05, | |
| "loss": 0.2074, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.793798449612403, | |
| "grad_norm": 3.686628580093384, | |
| "learning_rate": 1.9515392637600583e-05, | |
| "loss": 0.2082, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.793798449612403, | |
| "eval_accuracy": 0.9262777023971054, | |
| "eval_f1": 0.926117263289722, | |
| "eval_loss": 0.17793285846710205, | |
| "eval_precision": 0.9261011200078554, | |
| "eval_recall": 0.9262777023971054, | |
| "eval_runtime": 31.9195, | |
| "eval_samples_per_second": 69.268, | |
| "eval_steps_per_second": 8.678, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.09922480620155, | |
| "grad_norm": 3.4548275470733643, | |
| "learning_rate": 1.9418480436245487e-05, | |
| "loss": 0.1998, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.409302325581395, | |
| "grad_norm": 1.4032052755355835, | |
| "learning_rate": 1.931302715849967e-05, | |
| "loss": 0.1723, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.409302325581395, | |
| "eval_accuracy": 0.9308005427408412, | |
| "eval_f1": 0.9301945813993165, | |
| "eval_loss": 0.1693263202905655, | |
| "eval_precision": 0.9315485344818776, | |
| "eval_recall": 0.9308005427408412, | |
| "eval_runtime": 31.8606, | |
| "eval_samples_per_second": 69.396, | |
| "eval_steps_per_second": 8.694, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.7193798449612405, | |
| "grad_norm": 2.225787401199341, | |
| "learning_rate": 1.9199128433855277e-05, | |
| "loss": 0.1864, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.024806201550387, | |
| "grad_norm": 4.083457946777344, | |
| "learning_rate": 1.9076887550492877e-05, | |
| "loss": 0.1877, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.024806201550387, | |
| "eval_accuracy": 0.9380370872908186, | |
| "eval_f1": 0.9381736344846147, | |
| "eval_loss": 0.168083056807518, | |
| "eval_precision": 0.9384861550793825, | |
| "eval_recall": 0.9380370872908186, | |
| "eval_runtime": 32.0939, | |
| "eval_samples_per_second": 68.892, | |
| "eval_steps_per_second": 8.631, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.334883720930232, | |
| "grad_norm": 5.380185127258301, | |
| "learning_rate": 1.894641536161537e-05, | |
| "loss": 0.1662, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 8.644961240310078, | |
| "grad_norm": 2.990643262863159, | |
| "learning_rate": 1.8807830184921636e-05, | |
| "loss": 0.2, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.644961240310078, | |
| "eval_accuracy": 0.9402985074626866, | |
| "eval_f1": 0.9401756807643212, | |
| "eval_loss": 0.14824680984020233, | |
| "eval_precision": 0.9401845264674767, | |
| "eval_recall": 0.9402985074626866, | |
| "eval_runtime": 31.8892, | |
| "eval_samples_per_second": 69.334, | |
| "eval_steps_per_second": 8.686, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.955038759689922, | |
| "grad_norm": 3.6064834594726562, | |
| "learning_rate": 1.8661257695311006e-05, | |
| "loss": 0.1621, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.26046511627907, | |
| "grad_norm": 5.486879825592041, | |
| "learning_rate": 1.8506830810915983e-05, | |
| "loss": 0.1642, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.26046511627907, | |
| "eval_accuracy": 0.9330619629127092, | |
| "eval_f1": 0.9322023134638141, | |
| "eval_loss": 0.16374026238918304, | |
| "eval_precision": 0.9352048827585053, | |
| "eval_recall": 0.9330619629127092, | |
| "eval_runtime": 32.0395, | |
| "eval_samples_per_second": 69.009, | |
| "eval_steps_per_second": 8.646, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.570542635658915, | |
| "grad_norm": 2.025575637817383, | |
| "learning_rate": 1.8344689572566436e-05, | |
| "loss": 0.1586, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 9.88062015503876, | |
| "grad_norm": 3.0411391258239746, | |
| "learning_rate": 1.8174981016794653e-05, | |
| "loss": 0.1525, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.88062015503876, | |
| "eval_accuracy": 0.942107643600181, | |
| "eval_f1": 0.94173840230828, | |
| "eval_loss": 0.14938268065452576, | |
| "eval_precision": 0.9424881755892641, | |
| "eval_recall": 0.942107643600181, | |
| "eval_runtime": 31.8834, | |
| "eval_samples_per_second": 69.347, | |
| "eval_steps_per_second": 8.688, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.186046511627907, | |
| "grad_norm": 3.4836244583129883, | |
| "learning_rate": 1.799785904249642e-05, | |
| "loss": 0.1649, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 10.496124031007753, | |
| "grad_norm": 3.605983257293701, | |
| "learning_rate": 1.7813484271368947e-05, | |
| "loss": 0.158, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.496124031007753, | |
| "eval_accuracy": 0.9484396200814111, | |
| "eval_f1": 0.947995545399902, | |
| "eval_loss": 0.1402619332075119, | |
| "eval_precision": 0.9494961047542965, | |
| "eval_recall": 0.9484396200814111, | |
| "eval_runtime": 32.0182, | |
| "eval_samples_per_second": 69.055, | |
| "eval_steps_per_second": 8.651, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.806201550387597, | |
| "grad_norm": 4.386783123016357, | |
| "learning_rate": 1.7622023902252378e-05, | |
| "loss": 0.1669, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 11.111627906976745, | |
| "grad_norm": 3.84214186668396, | |
| "learning_rate": 1.742365155950676e-05, | |
| "loss": 0.1327, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.111627906976745, | |
| "eval_accuracy": 0.9497964721845319, | |
| "eval_f1": 0.9498129231725488, | |
| "eval_loss": 0.13292284309864044, | |
| "eval_precision": 0.9498330407980847, | |
| "eval_recall": 0.9497964721845319, | |
| "eval_runtime": 32.0593, | |
| "eval_samples_per_second": 68.966, | |
| "eval_steps_per_second": 8.64, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.421705426356588, | |
| "grad_norm": 2.3866167068481445, | |
| "learning_rate": 1.7218547135562155e-05, | |
| "loss": 0.1679, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 11.731782945736434, | |
| "grad_norm": 5.185602188110352, | |
| "learning_rate": 1.700689662778458e-05, | |
| "loss": 0.1465, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 11.731782945736434, | |
| "eval_accuracy": 0.9525101763907734, | |
| "eval_f1": 0.9523724534109135, | |
| "eval_loss": 0.12334821373224258, | |
| "eval_precision": 0.9524932751901988, | |
| "eval_recall": 0.9525101763907734, | |
| "eval_runtime": 31.92, | |
| "eval_samples_per_second": 69.267, | |
| "eval_steps_per_second": 8.678, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.037209302325582, | |
| "grad_norm": 9.162732124328613, | |
| "learning_rate": 1.678889196980572e-05, | |
| "loss": 0.1291, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 12.347286821705426, | |
| "grad_norm": 5.636198043823242, | |
| "learning_rate": 1.6564730857469392e-05, | |
| "loss": 0.1311, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.347286821705426, | |
| "eval_accuracy": 0.9520578923563998, | |
| "eval_f1": 0.9520149841974469, | |
| "eval_loss": 0.12798894941806793, | |
| "eval_precision": 0.9519983389426212, | |
| "eval_recall": 0.9520578923563998, | |
| "eval_runtime": 31.9866, | |
| "eval_samples_per_second": 69.123, | |
| "eval_steps_per_second": 8.66, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.657364341085271, | |
| "grad_norm": 2.2612485885620117, | |
| "learning_rate": 1.633461656955259e-05, | |
| "loss": 0.1619, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 12.967441860465117, | |
| "grad_norm": 2.522465229034424, | |
| "learning_rate": 1.609875778342372e-05, | |
| "loss": 0.129, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 12.967441860465117, | |
| "eval_accuracy": 0.9556761646313885, | |
| "eval_f1": 0.9555744118131845, | |
| "eval_loss": 0.11725445836782455, | |
| "eval_precision": 0.9556383737245242, | |
| "eval_recall": 0.9556761646313885, | |
| "eval_runtime": 31.8471, | |
| "eval_samples_per_second": 69.426, | |
| "eval_steps_per_second": 8.698, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.272868217054263, | |
| "grad_norm": 7.201235294342041, | |
| "learning_rate": 1.5857368385805087e-05, | |
| "loss": 0.1351, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 13.582945736434109, | |
| "grad_norm": 4.276857852935791, | |
| "learning_rate": 1.561565191643536e-05, | |
| "loss": 0.1425, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 13.582945736434109, | |
| "eval_accuracy": 0.9552238805970149, | |
| "eval_f1": 0.9552090145107874, | |
| "eval_loss": 0.11896734684705734, | |
| "eval_precision": 0.9551978476090869, | |
| "eval_recall": 0.9552238805970149, | |
| "eval_runtime": 31.9931, | |
| "eval_samples_per_second": 69.109, | |
| "eval_steps_per_second": 8.658, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 13.893023255813954, | |
| "grad_norm": 3.126692295074463, | |
| "learning_rate": 1.5363962356243304e-05, | |
| "loss": 0.1249, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 14.1984496124031, | |
| "grad_norm": 2.0163676738739014, | |
| "learning_rate": 1.5107408528131196e-05, | |
| "loss": 0.1256, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.1984496124031, | |
| "eval_accuracy": 0.9565807327001357, | |
| "eval_f1": 0.9563271746916711, | |
| "eval_loss": 0.12249390035867691, | |
| "eval_precision": 0.9570420286206239, | |
| "eval_recall": 0.9565807327001357, | |
| "eval_runtime": 32.0483, | |
| "eval_samples_per_second": 68.99, | |
| "eval_steps_per_second": 8.643, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.508527131782946, | |
| "grad_norm": 2.3576133251190186, | |
| "learning_rate": 1.4846223085960681e-05, | |
| "loss": 0.1331, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 14.81860465116279, | |
| "grad_norm": 3.005863666534424, | |
| "learning_rate": 1.45806428837369e-05, | |
| "loss": 0.1461, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 14.81860465116279, | |
| "eval_accuracy": 0.9588421528720036, | |
| "eval_f1": 0.9587720391887182, | |
| "eval_loss": 0.11707326769828796, | |
| "eval_precision": 0.9587952797993504, | |
| "eval_recall": 0.9588421528720036, | |
| "eval_runtime": 31.9123, | |
| "eval_samples_per_second": 69.284, | |
| "eval_steps_per_second": 8.68, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 15.124031007751938, | |
| "grad_norm": 2.8450655937194824, | |
| "learning_rate": 1.4310908760819285e-05, | |
| "loss": 0.1193, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 15.434108527131784, | |
| "grad_norm": 2.168471097946167, | |
| "learning_rate": 1.4037265323518223e-05, | |
| "loss": 0.133, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 15.434108527131784, | |
| "eval_accuracy": 0.9547715965626413, | |
| "eval_f1": 0.9545783302992185, | |
| "eval_loss": 0.11651919037103653, | |
| "eval_precision": 0.9549254114560721, | |
| "eval_recall": 0.9547715965626413, | |
| "eval_runtime": 31.9187, | |
| "eval_samples_per_second": 69.27, | |
| "eval_steps_per_second": 8.678, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 15.744186046511627, | |
| "grad_norm": 4.06069278717041, | |
| "learning_rate": 1.375996072327573e-05, | |
| "loss": 0.117, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 16.049612403100774, | |
| "grad_norm": 0.5793169736862183, | |
| "learning_rate": 1.3479246431631214e-05, | |
| "loss": 0.1258, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 16.049612403100774, | |
| "eval_accuracy": 0.9479873360470376, | |
| "eval_f1": 0.9473924145380525, | |
| "eval_loss": 0.1302204728126526, | |
| "eval_precision": 0.9500128100803557, | |
| "eval_recall": 0.9479873360470376, | |
| "eval_runtime": 32.0659, | |
| "eval_samples_per_second": 68.952, | |
| "eval_steps_per_second": 8.638, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 16.35968992248062, | |
| "grad_norm": 3.8317296504974365, | |
| "learning_rate": 1.3195377012176449e-05, | |
| "loss": 0.1479, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 16.669767441860465, | |
| "grad_norm": 3.2260377407073975, | |
| "learning_rate": 1.2908609889706538e-05, | |
| "loss": 0.115, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 16.669767441860465, | |
| "eval_accuracy": 0.9534147444595206, | |
| "eval_f1": 0.9536709713857593, | |
| "eval_loss": 0.1319713443517685, | |
| "eval_precision": 0.9552310659222302, | |
| "eval_recall": 0.9534147444595206, | |
| "eval_runtime": 31.9757, | |
| "eval_samples_per_second": 69.146, | |
| "eval_steps_per_second": 8.663, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 16.97984496124031, | |
| "grad_norm": 1.9483188390731812, | |
| "learning_rate": 1.2619205116776244e-05, | |
| "loss": 0.1269, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 17.28527131782946, | |
| "grad_norm": 4.348479270935059, | |
| "learning_rate": 1.2327425137873315e-05, | |
| "loss": 0.1134, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 17.28527131782946, | |
| "eval_accuracy": 0.9552238805970149, | |
| "eval_f1": 0.9548701351406266, | |
| "eval_loss": 0.117097869515419, | |
| "eval_precision": 0.9561949341655123, | |
| "eval_recall": 0.9552238805970149, | |
| "eval_runtime": 31.9852, | |
| "eval_samples_per_second": 69.126, | |
| "eval_steps_per_second": 8.66, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 17.5953488372093, | |
| "grad_norm": 5.097657203674316, | |
| "learning_rate": 1.2033534551422718e-05, | |
| "loss": 0.1069, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 17.905426356589146, | |
| "grad_norm": 2.5790622234344482, | |
| "learning_rate": 1.173779986983762e-05, | |
| "loss": 0.1069, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 17.905426356589146, | |
| "eval_accuracy": 0.9597467209407508, | |
| "eval_f1": 0.9595276897831222, | |
| "eval_loss": 0.11270730942487717, | |
| "eval_precision": 0.9601617456499858, | |
| "eval_recall": 0.9597467209407508, | |
| "eval_runtime": 32.0361, | |
| "eval_samples_per_second": 69.016, | |
| "eval_steps_per_second": 8.646, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 17.905426356589146, | |
| "step": 2900, | |
| "total_flos": 1.0232905043346113e+19, | |
| "train_loss": 0.2441097328580659, | |
| "train_runtime": 4508.9942, | |
| "train_samples_per_second": 41.182, | |
| "train_steps_per_second": 1.285 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5796, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 36, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0232905043346113e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |