| { | |
| "best_metric": 0.691096901131333, | |
| "best_model_checkpoint": "de-DE/wav2vec2-xls-r-300m/checkpoint-5040", | |
| "epoch": 30.0, | |
| "eval_steps": 100, | |
| "global_step": 5400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 1.4890042543411255, | |
| "learning_rate": 9.074074074074075e-06, | |
| "loss": 4.091, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 1.1139342784881592, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 4.0628, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.7727408409118652, | |
| "learning_rate": 2.7592592592592594e-05, | |
| "loss": 3.9799, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_f1": 0.00205206930041668, | |
| "eval_loss": 3.8087375164031982, | |
| "eval_runtime": 21.0392, | |
| "eval_samples_per_second": 96.629, | |
| "eval_steps_per_second": 1.521, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 1.1827894449234009, | |
| "learning_rate": 3.685185185185185e-05, | |
| "loss": 3.8573, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 1.3509910106658936, | |
| "learning_rate": 4.6111111111111115e-05, | |
| "loss": 3.7732, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 1.5033173561096191, | |
| "learning_rate": 5.537037037037037e-05, | |
| "loss": 3.7526, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 1.2770198583602905, | |
| "learning_rate": 6.462962962962962e-05, | |
| "loss": 3.7773, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_f1": 0.00205206930041668, | |
| "eval_loss": 3.736783504486084, | |
| "eval_runtime": 21.1812, | |
| "eval_samples_per_second": 95.982, | |
| "eval_steps_per_second": 1.511, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 1.2693891525268555, | |
| "learning_rate": 7.38888888888889e-05, | |
| "loss": 3.7473, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.4083213806152344, | |
| "learning_rate": 8.314814814814815e-05, | |
| "loss": 3.7626, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 1.625786542892456, | |
| "learning_rate": 9.240740740740741e-05, | |
| "loss": 3.7632, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_f1": 0.00205206930041668, | |
| "eval_loss": 3.734524965286255, | |
| "eval_runtime": 21.1901, | |
| "eval_samples_per_second": 95.941, | |
| "eval_steps_per_second": 1.51, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.0555555555555554, | |
| "grad_norm": 1.2629117965698242, | |
| "learning_rate": 9.981481481481481e-05, | |
| "loss": 3.7604, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 1.164554476737976, | |
| "learning_rate": 9.878600823045268e-05, | |
| "loss": 3.7564, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.611111111111111, | |
| "grad_norm": 0.9783994555473328, | |
| "learning_rate": 9.775720164609054e-05, | |
| "loss": 3.7596, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 1.4372780323028564, | |
| "learning_rate": 9.672839506172841e-05, | |
| "loss": 3.7706, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.06443679291687161, | |
| "eval_f1": 0.00205206930041668, | |
| "eval_loss": 3.737985372543335, | |
| "eval_runtime": 20.8407, | |
| "eval_samples_per_second": 97.55, | |
| "eval_steps_per_second": 1.535, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 1.2122852802276611, | |
| "learning_rate": 9.569958847736626e-05, | |
| "loss": 3.7516, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 1.3332332372665405, | |
| "learning_rate": 9.467078189300412e-05, | |
| "loss": 3.735, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.722222222222222, | |
| "grad_norm": 0.9196431040763855, | |
| "learning_rate": 9.364197530864199e-05, | |
| "loss": 3.7224, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.3251221179962158, | |
| "learning_rate": 9.261316872427984e-05, | |
| "loss": 3.7518, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.09296606000983768, | |
| "eval_f1": 0.0059176939246660745, | |
| "eval_loss": 3.6887905597686768, | |
| "eval_runtime": 21.2519, | |
| "eval_samples_per_second": 95.662, | |
| "eval_steps_per_second": 1.506, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.277777777777778, | |
| "grad_norm": 1.470018982887268, | |
| "learning_rate": 9.15843621399177e-05, | |
| "loss": 3.7136, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 1.4413223266601562, | |
| "learning_rate": 9.055555555555556e-05, | |
| "loss": 3.6818, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 1.2150436639785767, | |
| "learning_rate": 8.952674897119342e-05, | |
| "loss": 3.699, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.09936055090998525, | |
| "eval_f1": 0.0073766292774243535, | |
| "eval_loss": 3.6655001640319824, | |
| "eval_runtime": 20.9819, | |
| "eval_samples_per_second": 96.893, | |
| "eval_steps_per_second": 1.525, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 6.111111111111111, | |
| "grad_norm": 1.4290884733200073, | |
| "learning_rate": 8.849794238683128e-05, | |
| "loss": 3.6581, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.388888888888889, | |
| "grad_norm": 1.6972917318344116, | |
| "learning_rate": 8.746913580246914e-05, | |
| "loss": 3.6401, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 1.3264811038970947, | |
| "learning_rate": 8.644032921810699e-05, | |
| "loss": 3.6843, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 1.6939361095428467, | |
| "learning_rate": 8.541152263374486e-05, | |
| "loss": 3.6017, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.11116576487948844, | |
| "eval_f1": 0.007963237845435386, | |
| "eval_loss": 3.613088846206665, | |
| "eval_runtime": 21.1422, | |
| "eval_samples_per_second": 96.158, | |
| "eval_steps_per_second": 1.514, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 7.222222222222222, | |
| "grad_norm": 1.3342170715332031, | |
| "learning_rate": 8.438271604938272e-05, | |
| "loss": 3.5671, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 2.0039055347442627, | |
| "learning_rate": 8.335390946502059e-05, | |
| "loss": 3.563, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 7.777777777777778, | |
| "grad_norm": 1.9838131666183472, | |
| "learning_rate": 8.232510288065844e-05, | |
| "loss": 3.4976, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.12297097884899164, | |
| "eval_f1": 0.0207489285319609, | |
| "eval_loss": 3.5292012691497803, | |
| "eval_runtime": 21.1877, | |
| "eval_samples_per_second": 95.952, | |
| "eval_steps_per_second": 1.51, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 8.055555555555555, | |
| "grad_norm": 1.7923851013183594, | |
| "learning_rate": 8.12962962962963e-05, | |
| "loss": 3.5203, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 2.5529909133911133, | |
| "learning_rate": 8.026748971193417e-05, | |
| "loss": 3.4522, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.61111111111111, | |
| "grad_norm": 1.8508727550506592, | |
| "learning_rate": 7.925925925925926e-05, | |
| "loss": 3.418, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 2.031550407409668, | |
| "learning_rate": 7.823045267489712e-05, | |
| "loss": 3.379, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.14805705853418594, | |
| "eval_f1": 0.030926614730120143, | |
| "eval_loss": 3.388072967529297, | |
| "eval_runtime": 21.1544, | |
| "eval_samples_per_second": 96.103, | |
| "eval_steps_per_second": 1.513, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 2.7599411010742188, | |
| "learning_rate": 7.720164609053499e-05, | |
| "loss": 3.3362, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 9.444444444444445, | |
| "grad_norm": 3.2844510078430176, | |
| "learning_rate": 7.617283950617285e-05, | |
| "loss": 3.3089, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.722222222222221, | |
| "grad_norm": 3.1966094970703125, | |
| "learning_rate": 7.516460905349794e-05, | |
| "loss": 3.2753, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.316312074661255, | |
| "learning_rate": 7.413580246913581e-05, | |
| "loss": 3.2182, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.19527791441219872, | |
| "eval_f1": 0.04058892684496088, | |
| "eval_loss": 3.1913669109344482, | |
| "eval_runtime": 20.9617, | |
| "eval_samples_per_second": 96.986, | |
| "eval_steps_per_second": 1.527, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.277777777777779, | |
| "grad_norm": 3.078585624694824, | |
| "learning_rate": 7.310699588477367e-05, | |
| "loss": 3.0993, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 10.555555555555555, | |
| "grad_norm": 6.923484802246094, | |
| "learning_rate": 7.207818930041152e-05, | |
| "loss": 3.0807, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.833333333333334, | |
| "grad_norm": 5.272449970245361, | |
| "learning_rate": 7.104938271604938e-05, | |
| "loss": 2.9804, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.2267584849975406, | |
| "eval_f1": 0.05879115082822113, | |
| "eval_loss": 2.93943452835083, | |
| "eval_runtime": 21.2521, | |
| "eval_samples_per_second": 95.661, | |
| "eval_steps_per_second": 1.506, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 11.11111111111111, | |
| "grad_norm": 7.379478931427002, | |
| "learning_rate": 7.002057613168725e-05, | |
| "loss": 2.9518, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.38888888888889, | |
| "grad_norm": 4.347708225250244, | |
| "learning_rate": 6.89917695473251e-05, | |
| "loss": 2.8607, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 11.666666666666666, | |
| "grad_norm": 3.146164894104004, | |
| "learning_rate": 6.796296296296296e-05, | |
| "loss": 2.805, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.944444444444445, | |
| "grad_norm": 3.2119126319885254, | |
| "learning_rate": 6.693415637860082e-05, | |
| "loss": 2.7577, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.2838170191834727, | |
| "eval_f1": 0.09477911894276655, | |
| "eval_loss": 2.6934072971343994, | |
| "eval_runtime": 21.1897, | |
| "eval_samples_per_second": 95.943, | |
| "eval_steps_per_second": 1.51, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 12.222222222222221, | |
| "grad_norm": 5.8592329025268555, | |
| "learning_rate": 6.590534979423869e-05, | |
| "loss": 2.6594, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 3.185109853744507, | |
| "learning_rate": 6.487654320987656e-05, | |
| "loss": 2.5666, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 12.777777777777779, | |
| "grad_norm": 6.137824535369873, | |
| "learning_rate": 6.384773662551441e-05, | |
| "loss": 2.5161, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.3320216428922774, | |
| "eval_f1": 0.13001894064725447, | |
| "eval_loss": 2.445955753326416, | |
| "eval_runtime": 21.1936, | |
| "eval_samples_per_second": 95.925, | |
| "eval_steps_per_second": 1.51, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 13.055555555555555, | |
| "grad_norm": 3.701847791671753, | |
| "learning_rate": 6.281893004115227e-05, | |
| "loss": 2.408, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 3.357701539993286, | |
| "learning_rate": 6.179012345679012e-05, | |
| "loss": 2.2915, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.61111111111111, | |
| "grad_norm": 4.91466760635376, | |
| "learning_rate": 6.0761316872427985e-05, | |
| "loss": 2.2576, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 13.88888888888889, | |
| "grad_norm": 4.737729072570801, | |
| "learning_rate": 5.973251028806585e-05, | |
| "loss": 2.2414, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.3890801770782095, | |
| "eval_f1": 0.1854455337191276, | |
| "eval_loss": 2.230682373046875, | |
| "eval_runtime": 21.2287, | |
| "eval_samples_per_second": 95.767, | |
| "eval_steps_per_second": 1.507, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 14.166666666666666, | |
| "grad_norm": 12.414363861083984, | |
| "learning_rate": 5.8703703703703704e-05, | |
| "loss": 2.125, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 14.444444444444445, | |
| "grad_norm": 4.032175540924072, | |
| "learning_rate": 5.7674897119341566e-05, | |
| "loss": 2.0399, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.722222222222221, | |
| "grad_norm": 5.251100540161133, | |
| "learning_rate": 5.664609053497942e-05, | |
| "loss": 2.023, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.966403484344482, | |
| "learning_rate": 5.5617283950617284e-05, | |
| "loss": 1.944, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.45204131824889326, | |
| "eval_f1": 0.2558497963614674, | |
| "eval_loss": 2.048044204711914, | |
| "eval_runtime": 21.1364, | |
| "eval_samples_per_second": 96.185, | |
| "eval_steps_per_second": 1.514, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 15.277777777777779, | |
| "grad_norm": 4.946567535400391, | |
| "learning_rate": 5.458847736625514e-05, | |
| "loss": 1.7941, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 15.555555555555555, | |
| "grad_norm": 6.810694217681885, | |
| "learning_rate": 5.355967078189301e-05, | |
| "loss": 1.7366, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.833333333333334, | |
| "grad_norm": 7.022438049316406, | |
| "learning_rate": 5.253086419753087e-05, | |
| "loss": 1.7764, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5081160846040335, | |
| "eval_f1": 0.32205117403665584, | |
| "eval_loss": 1.9176976680755615, | |
| "eval_runtime": 21.2753, | |
| "eval_samples_per_second": 95.557, | |
| "eval_steps_per_second": 1.504, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 16.11111111111111, | |
| "grad_norm": 4.259357929229736, | |
| "learning_rate": 5.150205761316873e-05, | |
| "loss": 1.6464, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 16.38888888888889, | |
| "grad_norm": 3.2829339504241943, | |
| "learning_rate": 5.047325102880659e-05, | |
| "loss": 1.5819, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 5.9115986824035645, | |
| "learning_rate": 4.9444444444444446e-05, | |
| "loss": 1.5292, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.944444444444443, | |
| "grad_norm": 5.184513092041016, | |
| "learning_rate": 4.841563786008231e-05, | |
| "loss": 1.5396, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5612395474667978, | |
| "eval_f1": 0.38352153766273633, | |
| "eval_loss": 1.7599705457687378, | |
| "eval_runtime": 21.2693, | |
| "eval_samples_per_second": 95.584, | |
| "eval_steps_per_second": 1.505, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 17.22222222222222, | |
| "grad_norm": 7.147132396697998, | |
| "learning_rate": 4.7386831275720164e-05, | |
| "loss": 1.3781, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 4.889599800109863, | |
| "learning_rate": 4.635802469135803e-05, | |
| "loss": 1.3611, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 17.77777777777778, | |
| "grad_norm": 4.039378643035889, | |
| "learning_rate": 4.532921810699588e-05, | |
| "loss": 1.3245, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6040334481062469, | |
| "eval_f1": 0.44146839497484525, | |
| "eval_loss": 1.6344797611236572, | |
| "eval_runtime": 21.2985, | |
| "eval_samples_per_second": 95.453, | |
| "eval_steps_per_second": 1.502, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 18.055555555555557, | |
| "grad_norm": 7.149341583251953, | |
| "learning_rate": 4.4300411522633745e-05, | |
| "loss": 1.3637, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 18.333333333333332, | |
| "grad_norm": 8.064897537231445, | |
| "learning_rate": 4.327160493827161e-05, | |
| "loss": 1.2293, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 18.61111111111111, | |
| "grad_norm": 7.903775691986084, | |
| "learning_rate": 4.224279835390947e-05, | |
| "loss": 1.2622, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 18.88888888888889, | |
| "grad_norm": 4.111396312713623, | |
| "learning_rate": 4.1213991769547326e-05, | |
| "loss": 1.1825, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.6246925725528775, | |
| "eval_f1": 0.4763520182391385, | |
| "eval_loss": 1.5767754316329956, | |
| "eval_runtime": 21.0448, | |
| "eval_samples_per_second": 96.603, | |
| "eval_steps_per_second": 1.521, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 19.166666666666668, | |
| "grad_norm": 6.990283966064453, | |
| "learning_rate": 4.018518518518519e-05, | |
| "loss": 1.1418, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 19.444444444444443, | |
| "grad_norm": 5.295820236206055, | |
| "learning_rate": 3.9156378600823044e-05, | |
| "loss": 1.1469, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 19.72222222222222, | |
| "grad_norm": 4.929720401763916, | |
| "learning_rate": 3.812757201646091e-05, | |
| "loss": 1.0467, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 9.373787879943848, | |
| "learning_rate": 3.709876543209877e-05, | |
| "loss": 1.0525, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6374815543531727, | |
| "eval_f1": 0.49386444314380973, | |
| "eval_loss": 1.5208549499511719, | |
| "eval_runtime": 21.2234, | |
| "eval_samples_per_second": 95.791, | |
| "eval_steps_per_second": 1.508, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 20.27777777777778, | |
| "grad_norm": 7.043643474578857, | |
| "learning_rate": 3.606995884773663e-05, | |
| "loss": 0.9412, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 20.555555555555557, | |
| "grad_norm": 4.172306537628174, | |
| "learning_rate": 3.504115226337449e-05, | |
| "loss": 0.9267, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.833333333333332, | |
| "grad_norm": 7.322742462158203, | |
| "learning_rate": 3.401234567901235e-05, | |
| "loss": 0.9496, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.6586325627151992, | |
| "eval_f1": 0.526846190918778, | |
| "eval_loss": 1.4922317266464233, | |
| "eval_runtime": 21.2176, | |
| "eval_samples_per_second": 95.816, | |
| "eval_steps_per_second": 1.508, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 21.11111111111111, | |
| "grad_norm": 3.450895071029663, | |
| "learning_rate": 3.2983539094650206e-05, | |
| "loss": 0.8906, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 21.38888888888889, | |
| "grad_norm": 4.777499675750732, | |
| "learning_rate": 3.195473251028807e-05, | |
| "loss": 0.895, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 21.666666666666668, | |
| "grad_norm": 6.931682109832764, | |
| "learning_rate": 3.0925925925925924e-05, | |
| "loss": 0.8423, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.944444444444443, | |
| "grad_norm": 4.988988876342773, | |
| "learning_rate": 2.989711934156379e-05, | |
| "loss": 0.8512, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6714215445154943, | |
| "eval_f1": 0.5555183831392889, | |
| "eval_loss": 1.4699825048446655, | |
| "eval_runtime": 20.1301, | |
| "eval_samples_per_second": 100.993, | |
| "eval_steps_per_second": 1.59, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 5.021641731262207, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.8724, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 12.957953453063965, | |
| "learning_rate": 2.786008230452675e-05, | |
| "loss": 0.7648, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 22.77777777777778, | |
| "grad_norm": 11.249091148376465, | |
| "learning_rate": 2.683127572016461e-05, | |
| "loss": 0.7718, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.6709296606000984, | |
| "eval_f1": 0.5596452801153066, | |
| "eval_loss": 1.5005793571472168, | |
| "eval_runtime": 20.1506, | |
| "eval_samples_per_second": 100.89, | |
| "eval_steps_per_second": 1.588, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 23.055555555555557, | |
| "grad_norm": 6.9278788566589355, | |
| "learning_rate": 2.580246913580247e-05, | |
| "loss": 0.7561, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 23.333333333333332, | |
| "grad_norm": 6.076067924499512, | |
| "learning_rate": 2.4773662551440328e-05, | |
| "loss": 0.6688, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 23.61111111111111, | |
| "grad_norm": 9.00234317779541, | |
| "learning_rate": 2.374485596707819e-05, | |
| "loss": 0.6977, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 23.88888888888889, | |
| "grad_norm": 6.011639595031738, | |
| "learning_rate": 2.271604938271605e-05, | |
| "loss": 0.7335, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6783079193310378, | |
| "eval_f1": 0.5742235106898269, | |
| "eval_loss": 1.4700865745544434, | |
| "eval_runtime": 20.1667, | |
| "eval_samples_per_second": 100.81, | |
| "eval_steps_per_second": 1.587, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 24.166666666666668, | |
| "grad_norm": 6.378417491912842, | |
| "learning_rate": 2.168724279835391e-05, | |
| "loss": 0.7581, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 24.444444444444443, | |
| "grad_norm": 5.407553672790527, | |
| "learning_rate": 2.0658436213991768e-05, | |
| "loss": 0.645, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 24.72222222222222, | |
| "grad_norm": 4.054404258728027, | |
| "learning_rate": 1.962962962962963e-05, | |
| "loss": 0.644, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 6.489627838134766, | |
| "learning_rate": 1.860082304526749e-05, | |
| "loss": 0.6514, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.6787998032464339, | |
| "eval_f1": 0.5776489069621243, | |
| "eval_loss": 1.5083099603652954, | |
| "eval_runtime": 20.1289, | |
| "eval_samples_per_second": 100.999, | |
| "eval_steps_per_second": 1.59, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.27777777777778, | |
| "grad_norm": 4.55994176864624, | |
| "learning_rate": 1.757201646090535e-05, | |
| "loss": 0.6414, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 25.555555555555557, | |
| "grad_norm": 3.7992448806762695, | |
| "learning_rate": 1.654320987654321e-05, | |
| "loss": 0.572, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 25.833333333333332, | |
| "grad_norm": 28.1027774810791, | |
| "learning_rate": 1.551440329218107e-05, | |
| "loss": 0.6161, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.6802754549926218, | |
| "eval_f1": 0.5875122879521407, | |
| "eval_loss": 1.532254695892334, | |
| "eval_runtime": 20.1215, | |
| "eval_samples_per_second": 101.036, | |
| "eval_steps_per_second": 1.59, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 26.11111111111111, | |
| "grad_norm": 6.135079383850098, | |
| "learning_rate": 1.448559670781893e-05, | |
| "loss": 0.6043, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 26.38888888888889, | |
| "grad_norm": 4.171236515045166, | |
| "learning_rate": 1.3477366255144033e-05, | |
| "loss": 0.6035, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 5.243923664093018, | |
| "learning_rate": 1.2448559670781894e-05, | |
| "loss": 0.5319, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 26.944444444444443, | |
| "grad_norm": 7.352174282073975, | |
| "learning_rate": 1.1419753086419753e-05, | |
| "loss": 0.5576, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.6822429906542056, | |
| "eval_f1": 0.5893952239065103, | |
| "eval_loss": 1.5246381759643555, | |
| "eval_runtime": 20.121, | |
| "eval_samples_per_second": 101.039, | |
| "eval_steps_per_second": 1.59, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 27.22222222222222, | |
| "grad_norm": 6.190335750579834, | |
| "learning_rate": 1.0390946502057614e-05, | |
| "loss": 0.5641, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "grad_norm": 4.43634557723999, | |
| "learning_rate": 9.362139917695475e-06, | |
| "loss": 0.5445, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 27.77777777777778, | |
| "grad_norm": 3.3575663566589355, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.5092, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.691096901131333, | |
| "eval_f1": 0.6042086142382966, | |
| "eval_loss": 1.5102460384368896, | |
| "eval_runtime": 20.1291, | |
| "eval_samples_per_second": 100.998, | |
| "eval_steps_per_second": 1.59, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 28.055555555555557, | |
| "grad_norm": 11.934253692626953, | |
| "learning_rate": 7.304526748971194e-06, | |
| "loss": 0.5371, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 28.333333333333332, | |
| "grad_norm": 5.053288459777832, | |
| "learning_rate": 6.275720164609055e-06, | |
| "loss": 0.498, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 28.61111111111111, | |
| "grad_norm": 4.996842861175537, | |
| "learning_rate": 5.246913580246914e-06, | |
| "loss": 0.5505, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 28.88888888888889, | |
| "grad_norm": 13.71835994720459, | |
| "learning_rate": 4.218106995884774e-06, | |
| "loss": 0.5109, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.6886374815543532, | |
| "eval_f1": 0.600059116410086, | |
| "eval_loss": 1.5202569961547852, | |
| "eval_runtime": 20.2693, | |
| "eval_samples_per_second": 100.299, | |
| "eval_steps_per_second": 1.579, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 29.166666666666668, | |
| "grad_norm": 3.211416721343994, | |
| "learning_rate": 3.1893004115226337e-06, | |
| "loss": 0.5093, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 29.444444444444443, | |
| "grad_norm": 4.839755058288574, | |
| "learning_rate": 2.1604938271604937e-06, | |
| "loss": 0.5158, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 29.72222222222222, | |
| "grad_norm": 5.453362464904785, | |
| "learning_rate": 1.131687242798354e-06, | |
| "loss": 0.522, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 2.713297128677368, | |
| "learning_rate": 1.0288065843621401e-07, | |
| "loss": 0.5301, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.690605017215937, | |
| "eval_f1": 0.603264981616179, | |
| "eval_loss": 1.5189286470413208, | |
| "eval_runtime": 20.2156, | |
| "eval_samples_per_second": 100.566, | |
| "eval_steps_per_second": 1.583, | |
| "step": 5400 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 1.0469261495692793e+20, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |