| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 112400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 39.0721435546875, | |
| "learning_rate": 9.800622775800713e-06, | |
| "loss": 1.2025, | |
| "step": 2248 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5195729732513428, | |
| "eval_loss": 1.3446124792099, | |
| "eval_runtime": 43.8683, | |
| "eval_samples_per_second": 12.811, | |
| "eval_steps_per_second": 12.811, | |
| "step": 2248 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.9895684123039246, | |
| "learning_rate": 9.600800711743772e-06, | |
| "loss": 1.415, | |
| "step": 4496 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6032028198242188, | |
| "eval_loss": 1.6350069046020508, | |
| "eval_runtime": 43.9062, | |
| "eval_samples_per_second": 12.8, | |
| "eval_steps_per_second": 12.8, | |
| "step": 4496 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.6042933464050293, | |
| "learning_rate": 9.400800711743772e-06, | |
| "loss": 1.4176, | |
| "step": 6744 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6583629846572876, | |
| "eval_loss": 1.6250243186950684, | |
| "eval_runtime": 44.433, | |
| "eval_samples_per_second": 12.648, | |
| "eval_steps_per_second": 12.648, | |
| "step": 6744 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.39769357442855835, | |
| "learning_rate": 9.200889679715304e-06, | |
| "loss": 1.384, | |
| "step": 8992 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7241992950439453, | |
| "eval_loss": 1.3693625926971436, | |
| "eval_runtime": 44.1124, | |
| "eval_samples_per_second": 12.74, | |
| "eval_steps_per_second": 12.74, | |
| "step": 8992 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 18.08792495727539, | |
| "learning_rate": 9.000978647686834e-06, | |
| "loss": 1.3658, | |
| "step": 11240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.709964394569397, | |
| "eval_loss": 1.4330979585647583, | |
| "eval_runtime": 44.4448, | |
| "eval_samples_per_second": 12.645, | |
| "eval_steps_per_second": 12.645, | |
| "step": 11240 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.2233850061893463, | |
| "learning_rate": 8.800978647686834e-06, | |
| "loss": 1.2763, | |
| "step": 13488 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7437722682952881, | |
| "eval_loss": 1.3310519456863403, | |
| "eval_runtime": 45.1045, | |
| "eval_samples_per_second": 12.46, | |
| "eval_steps_per_second": 12.46, | |
| "step": 13488 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 27.293367385864258, | |
| "learning_rate": 8.601156583629893e-06, | |
| "loss": 1.2175, | |
| "step": 15736 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7580071091651917, | |
| "eval_loss": 1.2726552486419678, | |
| "eval_runtime": 44.7638, | |
| "eval_samples_per_second": 12.555, | |
| "eval_steps_per_second": 12.555, | |
| "step": 15736 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.855543613433838, | |
| "learning_rate": 8.401156583629893e-06, | |
| "loss": 1.1276, | |
| "step": 17984 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7330960631370544, | |
| "eval_loss": 1.4520480632781982, | |
| "eval_runtime": 44.2783, | |
| "eval_samples_per_second": 12.692, | |
| "eval_steps_per_second": 12.692, | |
| "step": 17984 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.08200906962156296, | |
| "learning_rate": 8.201245551601425e-06, | |
| "loss": 1.1053, | |
| "step": 20232 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.77224200963974, | |
| "eval_loss": 1.2134090662002563, | |
| "eval_runtime": 44.2632, | |
| "eval_samples_per_second": 12.697, | |
| "eval_steps_per_second": 12.697, | |
| "step": 20232 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.09874199330806732, | |
| "learning_rate": 8.001334519572955e-06, | |
| "loss": 1.0314, | |
| "step": 22480 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7829181551933289, | |
| "eval_loss": 1.2143168449401855, | |
| "eval_runtime": 44.2853, | |
| "eval_samples_per_second": 12.69, | |
| "eval_steps_per_second": 12.69, | |
| "step": 22480 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.23497702181339264, | |
| "learning_rate": 7.801423487544484e-06, | |
| "loss": 1.0029, | |
| "step": 24728 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7811387777328491, | |
| "eval_loss": 1.3311693668365479, | |
| "eval_runtime": 44.0086, | |
| "eval_samples_per_second": 12.77, | |
| "eval_steps_per_second": 12.77, | |
| "step": 24728 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.08876941353082657, | |
| "learning_rate": 7.6014234875444846e-06, | |
| "loss": 0.9108, | |
| "step": 26976 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8024911284446716, | |
| "eval_loss": 1.2227764129638672, | |
| "eval_runtime": 44.4307, | |
| "eval_samples_per_second": 12.649, | |
| "eval_steps_per_second": 12.649, | |
| "step": 26976 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.13910023868083954, | |
| "learning_rate": 7.401512455516014e-06, | |
| "loss": 0.8335, | |
| "step": 29224 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8078292012214661, | |
| "eval_loss": 1.1526196002960205, | |
| "eval_runtime": 44.5737, | |
| "eval_samples_per_second": 12.608, | |
| "eval_steps_per_second": 12.608, | |
| "step": 29224 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 159.14675903320312, | |
| "learning_rate": 7.201601423487545e-06, | |
| "loss": 0.8514, | |
| "step": 31472 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8202847242355347, | |
| "eval_loss": 0.9904452562332153, | |
| "eval_runtime": 44.2371, | |
| "eval_samples_per_second": 12.704, | |
| "eval_steps_per_second": 12.704, | |
| "step": 31472 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.09662589430809021, | |
| "learning_rate": 7.001601423487545e-06, | |
| "loss": 0.7389, | |
| "step": 33720 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8024911284446716, | |
| "eval_loss": 1.2999956607818604, | |
| "eval_runtime": 43.9303, | |
| "eval_samples_per_second": 12.793, | |
| "eval_steps_per_second": 12.793, | |
| "step": 33720 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.1123761385679245, | |
| "learning_rate": 6.801690391459075e-06, | |
| "loss": 0.6993, | |
| "step": 35968 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8202847242355347, | |
| "eval_loss": 1.0872814655303955, | |
| "eval_runtime": 44.4779, | |
| "eval_samples_per_second": 12.636, | |
| "eval_steps_per_second": 12.636, | |
| "step": 35968 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.5431187748908997, | |
| "learning_rate": 6.601690391459076e-06, | |
| "loss": 0.6177, | |
| "step": 38216 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8327401876449585, | |
| "eval_loss": 1.0856190919876099, | |
| "eval_runtime": 44.047, | |
| "eval_samples_per_second": 12.759, | |
| "eval_steps_per_second": 12.759, | |
| "step": 38216 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.034104421734809875, | |
| "learning_rate": 6.401868327402135e-06, | |
| "loss": 0.641, | |
| "step": 40464 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7971529960632324, | |
| "eval_loss": 1.3224000930786133, | |
| "eval_runtime": 43.8821, | |
| "eval_samples_per_second": 12.807, | |
| "eval_steps_per_second": 12.807, | |
| "step": 40464 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.05449577793478966, | |
| "learning_rate": 6.201957295373666e-06, | |
| "loss": 0.611, | |
| "step": 42712 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8291814923286438, | |
| "eval_loss": 1.1800107955932617, | |
| "eval_runtime": 43.8606, | |
| "eval_samples_per_second": 12.813, | |
| "eval_steps_per_second": 12.813, | |
| "step": 42712 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 24.56753158569336, | |
| "learning_rate": 6.001957295373666e-06, | |
| "loss": 0.5744, | |
| "step": 44960 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.809608519077301, | |
| "eval_loss": 1.2937426567077637, | |
| "eval_runtime": 44.1614, | |
| "eval_samples_per_second": 12.726, | |
| "eval_steps_per_second": 12.726, | |
| "step": 44960 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.14437897503376007, | |
| "learning_rate": 5.802046263345196e-06, | |
| "loss": 0.5008, | |
| "step": 47208 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8416370153427124, | |
| "eval_loss": 1.1565003395080566, | |
| "eval_runtime": 44.2577, | |
| "eval_samples_per_second": 12.698, | |
| "eval_steps_per_second": 12.698, | |
| "step": 47208 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.044370945543050766, | |
| "learning_rate": 5.6021352313167265e-06, | |
| "loss": 0.4396, | |
| "step": 49456 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8149465918540955, | |
| "eval_loss": 1.366288661956787, | |
| "eval_runtime": 44.4773, | |
| "eval_samples_per_second": 12.636, | |
| "eval_steps_per_second": 12.636, | |
| "step": 49456 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.21999333798885345, | |
| "learning_rate": 5.402224199288256e-06, | |
| "loss": 0.4313, | |
| "step": 51704 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8220640420913696, | |
| "eval_loss": 1.3266714811325073, | |
| "eval_runtime": 43.9122, | |
| "eval_samples_per_second": 12.798, | |
| "eval_steps_per_second": 12.798, | |
| "step": 51704 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.037919577211141586, | |
| "learning_rate": 5.202313167259787e-06, | |
| "loss": 0.3954, | |
| "step": 53952 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8469750881195068, | |
| "eval_loss": 1.1824334859848022, | |
| "eval_runtime": 44.2008, | |
| "eval_samples_per_second": 12.715, | |
| "eval_steps_per_second": 12.715, | |
| "step": 53952 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 106.5145263671875, | |
| "learning_rate": 5.002491103202848e-06, | |
| "loss": 0.4217, | |
| "step": 56200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8042704463005066, | |
| "eval_loss": 1.5585525035858154, | |
| "eval_runtime": 43.9488, | |
| "eval_samples_per_second": 12.788, | |
| "eval_steps_per_second": 12.788, | |
| "step": 56200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.015882058069109917, | |
| "learning_rate": 4.802491103202847e-06, | |
| "loss": 0.3797, | |
| "step": 58448 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8523131608963013, | |
| "eval_loss": 1.1745574474334717, | |
| "eval_runtime": 43.9074, | |
| "eval_samples_per_second": 12.8, | |
| "eval_steps_per_second": 12.8, | |
| "step": 58448 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.03550243377685547, | |
| "learning_rate": 4.602580071174377e-06, | |
| "loss": 0.358, | |
| "step": 60696 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8451957106590271, | |
| "eval_loss": 1.1937276124954224, | |
| "eval_runtime": 43.8357, | |
| "eval_samples_per_second": 12.821, | |
| "eval_steps_per_second": 12.821, | |
| "step": 60696 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.009829429909586906, | |
| "learning_rate": 4.402669039145908e-06, | |
| "loss": 0.2963, | |
| "step": 62944 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8309608697891235, | |
| "eval_loss": 1.4036489725112915, | |
| "eval_runtime": 44.7776, | |
| "eval_samples_per_second": 12.551, | |
| "eval_steps_per_second": 12.551, | |
| "step": 62944 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": Infinity, | |
| "learning_rate": 4.202758007117438e-06, | |
| "loss": 0.3338, | |
| "step": 65192 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8505337834358215, | |
| "eval_loss": 1.3133819103240967, | |
| "eval_runtime": 43.9603, | |
| "eval_samples_per_second": 12.784, | |
| "eval_steps_per_second": 12.784, | |
| "step": 65192 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.012281532399356365, | |
| "learning_rate": 4.0027580071174384e-06, | |
| "loss": 0.2565, | |
| "step": 67440 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8345195651054382, | |
| "eval_loss": 1.480637788772583, | |
| "eval_runtime": 44.0574, | |
| "eval_samples_per_second": 12.756, | |
| "eval_steps_per_second": 12.756, | |
| "step": 67440 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 0.0023416499607264996, | |
| "learning_rate": 3.8028469750889686e-06, | |
| "loss": 0.2798, | |
| "step": 69688 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8309608697891235, | |
| "eval_loss": 1.5173381567001343, | |
| "eval_runtime": 44.0999, | |
| "eval_samples_per_second": 12.744, | |
| "eval_steps_per_second": 12.744, | |
| "step": 69688 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 5.919681072235107, | |
| "learning_rate": 3.602935943060499e-06, | |
| "loss": 0.2674, | |
| "step": 71936 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8131672739982605, | |
| "eval_loss": 1.5758373737335205, | |
| "eval_runtime": 44.4876, | |
| "eval_samples_per_second": 12.633, | |
| "eval_steps_per_second": 12.633, | |
| "step": 71936 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 0.0024983042385429144, | |
| "learning_rate": 3.4029359430604986e-06, | |
| "loss": 0.2334, | |
| "step": 74184 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8558719158172607, | |
| "eval_loss": 1.3400838375091553, | |
| "eval_runtime": 44.5806, | |
| "eval_samples_per_second": 12.606, | |
| "eval_steps_per_second": 12.606, | |
| "step": 74184 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 0.0015680283540859818, | |
| "learning_rate": 3.2030249110320288e-06, | |
| "loss": 0.2352, | |
| "step": 76432 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8469750881195068, | |
| "eval_loss": 1.2716737985610962, | |
| "eval_runtime": 44.3768, | |
| "eval_samples_per_second": 12.664, | |
| "eval_steps_per_second": 12.664, | |
| "step": 76432 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.0035891232546418905, | |
| "learning_rate": 3.003113879003559e-06, | |
| "loss": 0.2406, | |
| "step": 78680 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8256227970123291, | |
| "eval_loss": 1.6162846088409424, | |
| "eval_runtime": 44.1614, | |
| "eval_samples_per_second": 12.726, | |
| "eval_steps_per_second": 12.726, | |
| "step": 78680 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.003527622204273939, | |
| "learning_rate": 2.8032028469750896e-06, | |
| "loss": 0.2208, | |
| "step": 80928 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8505337834358215, | |
| "eval_loss": 1.3815460205078125, | |
| "eval_runtime": 44.6638, | |
| "eval_samples_per_second": 12.583, | |
| "eval_steps_per_second": 12.583, | |
| "step": 80928 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 0.0015290452865883708, | |
| "learning_rate": 2.6033807829181497e-06, | |
| "loss": 0.1796, | |
| "step": 83176 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8576512336730957, | |
| "eval_loss": 1.392934799194336, | |
| "eval_runtime": 44.8201, | |
| "eval_samples_per_second": 12.539, | |
| "eval_steps_per_second": 12.539, | |
| "step": 83176 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 0.00960911251604557, | |
| "learning_rate": 2.4033807829181495e-06, | |
| "loss": 0.2127, | |
| "step": 85424 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8274021148681641, | |
| "eval_loss": 1.5271013975143433, | |
| "eval_runtime": 44.255, | |
| "eval_samples_per_second": 12.699, | |
| "eval_steps_per_second": 12.699, | |
| "step": 85424 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "grad_norm": 0.004349031951278448, | |
| "learning_rate": 2.20355871886121e-06, | |
| "loss": 0.1748, | |
| "step": 87672 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8416370153427124, | |
| "eval_loss": 1.50688636302948, | |
| "eval_runtime": 44.0422, | |
| "eval_samples_per_second": 12.76, | |
| "eval_steps_per_second": 12.76, | |
| "step": 87672 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.0012758744414895773, | |
| "learning_rate": 2.0035587188612103e-06, | |
| "loss": 0.1612, | |
| "step": 89920 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8469750881195068, | |
| "eval_loss": 1.396645426750183, | |
| "eval_runtime": 44.1524, | |
| "eval_samples_per_second": 12.729, | |
| "eval_steps_per_second": 12.729, | |
| "step": 89920 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "grad_norm": 0.0022624109406024218, | |
| "learning_rate": 1.8036476868327405e-06, | |
| "loss": 0.1757, | |
| "step": 92168 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8469750881195068, | |
| "eval_loss": 1.4628039598464966, | |
| "eval_runtime": 44.3029, | |
| "eval_samples_per_second": 12.685, | |
| "eval_steps_per_second": 12.685, | |
| "step": 92168 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 0.007528578396886587, | |
| "learning_rate": 1.6037366548042707e-06, | |
| "loss": 0.1664, | |
| "step": 94416 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8523131608963013, | |
| "eval_loss": 1.3362900018692017, | |
| "eval_runtime": 43.78, | |
| "eval_samples_per_second": 12.837, | |
| "eval_steps_per_second": 12.837, | |
| "step": 94416 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "grad_norm": 0.0013060200726613402, | |
| "learning_rate": 1.4038256227758006e-06, | |
| "loss": 0.1313, | |
| "step": 96664 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8434163928031921, | |
| "eval_loss": 1.438825011253357, | |
| "eval_runtime": 43.9878, | |
| "eval_samples_per_second": 12.776, | |
| "eval_steps_per_second": 12.776, | |
| "step": 96664 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 0.042194243520498276, | |
| "learning_rate": 1.203914590747331e-06, | |
| "loss": 0.1272, | |
| "step": 98912 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8629893064498901, | |
| "eval_loss": 1.3669886589050293, | |
| "eval_runtime": 43.6968, | |
| "eval_samples_per_second": 12.861, | |
| "eval_steps_per_second": 12.861, | |
| "step": 98912 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.00658341683447361, | |
| "learning_rate": 1.0040035587188612e-06, | |
| "loss": 0.1127, | |
| "step": 101160 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.854092538356781, | |
| "eval_loss": 1.4244239330291748, | |
| "eval_runtime": 43.9697, | |
| "eval_samples_per_second": 12.782, | |
| "eval_steps_per_second": 12.782, | |
| "step": 101160 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "grad_norm": 0.0009845913155004382, | |
| "learning_rate": 8.040035587188613e-07, | |
| "loss": 0.1062, | |
| "step": 103408 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.854092538356781, | |
| "eval_loss": 1.3811708688735962, | |
| "eval_runtime": 44.3254, | |
| "eval_samples_per_second": 12.679, | |
| "eval_steps_per_second": 12.679, | |
| "step": 103408 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "grad_norm": 0.0018777468940243125, | |
| "learning_rate": 6.040925266903915e-07, | |
| "loss": 0.0924, | |
| "step": 105656 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.854092538356781, | |
| "eval_loss": 1.4448031187057495, | |
| "eval_runtime": 44.1197, | |
| "eval_samples_per_second": 12.738, | |
| "eval_steps_per_second": 12.738, | |
| "step": 105656 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.0018540391465649009, | |
| "learning_rate": 4.040925266903915e-07, | |
| "loss": 0.0998, | |
| "step": 107904 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8683273792266846, | |
| "eval_loss": 1.305090069770813, | |
| "eval_runtime": 43.7364, | |
| "eval_samples_per_second": 12.85, | |
| "eval_steps_per_second": 12.85, | |
| "step": 107904 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "grad_norm": 0.0024543912149965763, | |
| "learning_rate": 2.0418149466192174e-07, | |
| "loss": 0.1055, | |
| "step": 110152 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8701067566871643, | |
| "eval_loss": 1.262986660003662, | |
| "eval_runtime": 44.2858, | |
| "eval_samples_per_second": 12.69, | |
| "eval_steps_per_second": 12.69, | |
| "step": 110152 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.00034937995951622725, | |
| "learning_rate": 4.181494661921708e-09, | |
| "loss": 0.1073, | |
| "step": 112400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8629893064498901, | |
| "eval_loss": 1.268770456314087, | |
| "eval_runtime": 44.0645, | |
| "eval_samples_per_second": 12.754, | |
| "eval_steps_per_second": 12.754, | |
| "step": 112400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 112400, | |
| "total_flos": 1.024714436352e+19, | |
| "train_loss": 0.5453376764589357, | |
| "train_runtime": 19116.6154, | |
| "train_samples_per_second": 5.88, | |
| "train_steps_per_second": 5.88 | |
| } | |
| ], | |
| "logging_steps": 35, | |
| "max_steps": 112400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.024714436352e+19, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |