| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 36816, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9930309764864052e-05, | |
| "loss": 0.9224, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.7422312786551197, | |
| "eval_loss": 0.6256920099258423, | |
| "eval_runtime": 34.0513, | |
| "eval_samples_per_second": 288.242, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.965648750240075e-05, | |
| "loss": 0.6107, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.7959246051961284, | |
| "eval_loss": 0.5218016505241394, | |
| "eval_runtime": 34.3716, | |
| "eval_samples_per_second": 285.555, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.93821164979285e-05, | |
| "loss": 0.5464, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.8036678553234845, | |
| "eval_loss": 0.4963739216327667, | |
| "eval_runtime": 34.3352, | |
| "eval_samples_per_second": 285.858, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.9107745493456255e-05, | |
| "loss": 0.5153, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.8173204279164544, | |
| "eval_loss": 0.4657774269580841, | |
| "eval_runtime": 34.4768, | |
| "eval_samples_per_second": 284.684, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.8833374488984006e-05, | |
| "loss": 0.4934, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.8264900662251655, | |
| "eval_loss": 0.4544191062450409, | |
| "eval_runtime": 34.2913, | |
| "eval_samples_per_second": 286.224, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.8559003484511758e-05, | |
| "loss": 0.4863, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.817014773306164, | |
| "eval_loss": 0.469377726316452, | |
| "eval_runtime": 34.3798, | |
| "eval_samples_per_second": 285.488, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.828463248003951e-05, | |
| "loss": 0.4639, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.8337238920020377, | |
| "eval_loss": 0.44527676701545715, | |
| "eval_runtime": 34.3542, | |
| "eval_samples_per_second": 285.7, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.8010261475567264e-05, | |
| "loss": 0.477, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.83841059602649, | |
| "eval_loss": 0.4258698523044586, | |
| "eval_runtime": 34.3114, | |
| "eval_samples_per_second": 286.056, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.7735890471095016e-05, | |
| "loss": 0.4592, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.8421803362200713, | |
| "eval_loss": 0.4147617518901825, | |
| "eval_runtime": 34.3201, | |
| "eval_samples_per_second": 285.984, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.746151946662277e-05, | |
| "loss": 0.4449, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.8431991849210392, | |
| "eval_loss": 0.42169830203056335, | |
| "eval_runtime": 34.3613, | |
| "eval_samples_per_second": 285.641, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.718714846215052e-05, | |
| "loss": 0.4448, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.844319918492104, | |
| "eval_loss": 0.4075462818145752, | |
| "eval_runtime": 34.3101, | |
| "eval_samples_per_second": 286.067, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.6912777457678273e-05, | |
| "loss": 0.4416, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.8473764645950076, | |
| "eval_loss": 0.40427786111831665, | |
| "eval_runtime": 34.3011, | |
| "eval_samples_per_second": 286.142, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.6638406453206025e-05, | |
| "loss": 0.4396, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.845746306673459, | |
| "eval_loss": 0.40583938360214233, | |
| "eval_runtime": 34.3317, | |
| "eval_samples_per_second": 285.887, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.636403544873378e-05, | |
| "loss": 0.4262, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.8414671421293938, | |
| "eval_loss": 0.4056021273136139, | |
| "eval_runtime": 34.3275, | |
| "eval_samples_per_second": 285.923, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.609076192827942e-05, | |
| "loss": 0.4181, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8471726948548141, | |
| "eval_loss": 0.4094808101654053, | |
| "eval_runtime": 34.3454, | |
| "eval_samples_per_second": 285.773, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.5816390923807176e-05, | |
| "loss": 0.4277, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.8513499745287825, | |
| "eval_loss": 0.3897625207901001, | |
| "eval_runtime": 34.3305, | |
| "eval_samples_per_second": 285.898, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.5542019919334924e-05, | |
| "loss": 0.4204, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.8414671421293938, | |
| "eval_loss": 0.40381574630737305, | |
| "eval_runtime": 34.3185, | |
| "eval_samples_per_second": 285.997, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.526764891486268e-05, | |
| "loss": 0.408, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_accuracy": 0.8490066225165563, | |
| "eval_loss": 0.39596864581108093, | |
| "eval_runtime": 34.3274, | |
| "eval_samples_per_second": 285.923, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.4993277910390432e-05, | |
| "loss": 0.4147, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.8508405501782985, | |
| "eval_loss": 0.39388328790664673, | |
| "eval_runtime": 34.3118, | |
| "eval_samples_per_second": 286.053, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.4718906905918185e-05, | |
| "loss": 0.4087, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_accuracy": 0.8454406520631687, | |
| "eval_loss": 0.3990323841571808, | |
| "eval_runtime": 34.3039, | |
| "eval_samples_per_second": 286.119, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4444535901445937e-05, | |
| "loss": 0.4182, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.8559347936831381, | |
| "eval_loss": 0.37663206458091736, | |
| "eval_runtime": 34.3301, | |
| "eval_samples_per_second": 285.901, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.4170164896973688e-05, | |
| "loss": 0.3974, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.8542027508914927, | |
| "eval_loss": 0.38825666904449463, | |
| "eval_runtime": 34.3094, | |
| "eval_samples_per_second": 286.073, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.3895793892501441e-05, | |
| "loss": 0.4031, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.8604177279673968, | |
| "eval_loss": 0.372531533241272, | |
| "eval_runtime": 34.3046, | |
| "eval_samples_per_second": 286.113, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.3621422888029194e-05, | |
| "loss": 0.4012, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.8586856851757514, | |
| "eval_loss": 0.3715327978134155, | |
| "eval_runtime": 34.3023, | |
| "eval_samples_per_second": 286.133, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.334760062556589e-05, | |
| "loss": 0.3364, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.8536933265410087, | |
| "eval_loss": 0.4157598912715912, | |
| "eval_runtime": 34.3192, | |
| "eval_samples_per_second": 285.992, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.3073229621093644e-05, | |
| "loss": 0.256, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_accuracy": 0.8565461029037188, | |
| "eval_loss": 0.4385443329811096, | |
| "eval_runtime": 34.3274, | |
| "eval_samples_per_second": 285.923, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.2798858616621397e-05, | |
| "loss": 0.2682, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.8606214977075904, | |
| "eval_loss": 0.40176108479499817, | |
| "eval_runtime": 34.3252, | |
| "eval_samples_per_second": 285.941, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.252448761214915e-05, | |
| "loss": 0.2566, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.8628629648497198, | |
| "eval_loss": 0.40525978803634644, | |
| "eval_runtime": 34.3427, | |
| "eval_samples_per_second": 285.796, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.22501166076769e-05, | |
| "loss": 0.2672, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.8600101884870097, | |
| "eval_loss": 0.41904282569885254, | |
| "eval_runtime": 34.3153, | |
| "eval_samples_per_second": 286.024, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1975745603204653e-05, | |
| "loss": 0.2614, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.8596026490066225, | |
| "eval_loss": 0.4234951138496399, | |
| "eval_runtime": 34.3391, | |
| "eval_samples_per_second": 285.825, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.1701374598732406e-05, | |
| "loss": 0.2581, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.8562404482934284, | |
| "eval_loss": 0.4271596670150757, | |
| "eval_runtime": 34.3468, | |
| "eval_samples_per_second": 285.762, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.142700359426016e-05, | |
| "loss": 0.2623, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.8616403464085584, | |
| "eval_loss": 0.42012786865234375, | |
| "eval_runtime": 34.342, | |
| "eval_samples_per_second": 285.802, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.1153181331796856e-05, | |
| "loss": 0.2536, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.8611309220580744, | |
| "eval_loss": 0.42309054732322693, | |
| "eval_runtime": 34.3494, | |
| "eval_samples_per_second": 285.74, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.0878810327324609e-05, | |
| "loss": 0.2595, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.863779928680591, | |
| "eval_loss": 0.408088356256485, | |
| "eval_runtime": 34.2723, | |
| "eval_samples_per_second": 286.383, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.0604439322852362e-05, | |
| "loss": 0.2626, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.8558329088130413, | |
| "eval_loss": 0.4161333441734314, | |
| "eval_runtime": 34.2334, | |
| "eval_samples_per_second": 286.708, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.0330068318380115e-05, | |
| "loss": 0.2607, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.8580743759551707, | |
| "eval_loss": 0.42380374670028687, | |
| "eval_runtime": 34.2608, | |
| "eval_samples_per_second": 286.479, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.0055697313907867e-05, | |
| "loss": 0.268, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_accuracy": 0.8599083036169128, | |
| "eval_loss": 0.43461742997169495, | |
| "eval_runtime": 34.234, | |
| "eval_samples_per_second": 286.703, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.78132630943562e-06, | |
| "loss": 0.2702, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_accuracy": 0.8628629648497198, | |
| "eval_loss": 0.4047214388847351, | |
| "eval_runtime": 34.2365, | |
| "eval_samples_per_second": 286.682, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 9.506955304963372e-06, | |
| "loss": 0.2496, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.8595007641365258, | |
| "eval_loss": 0.4368164539337158, | |
| "eval_runtime": 34.2332, | |
| "eval_samples_per_second": 286.71, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.232584300491125e-06, | |
| "loss": 0.2681, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.867142129393785, | |
| "eval_loss": 0.400213360786438, | |
| "eval_runtime": 34.2373, | |
| "eval_samples_per_second": 286.675, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.958762038027821e-06, | |
| "loss": 0.2608, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_accuracy": 0.863779928680591, | |
| "eval_loss": 0.40518832206726074, | |
| "eval_runtime": 34.2681, | |
| "eval_samples_per_second": 286.418, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.684939775564519e-06, | |
| "loss": 0.261, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.8655119714722364, | |
| "eval_loss": 0.3955936133861542, | |
| "eval_runtime": 34.2769, | |
| "eval_samples_per_second": 286.344, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.41056877109227e-06, | |
| "loss": 0.267, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.8610290371879776, | |
| "eval_loss": 0.40947309136390686, | |
| "eval_runtime": 34.2634, | |
| "eval_samples_per_second": 286.457, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.136197766620024e-06, | |
| "loss": 0.2598, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.8642893530310749, | |
| "eval_loss": 0.38150227069854736, | |
| "eval_runtime": 34.2625, | |
| "eval_samples_per_second": 286.465, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.861826762147777e-06, | |
| "loss": 0.2577, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.8641874681609781, | |
| "eval_loss": 0.401090145111084, | |
| "eval_runtime": 34.2614, | |
| "eval_samples_per_second": 286.474, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 7.587455757675529e-06, | |
| "loss": 0.2583, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.8678553234844626, | |
| "eval_loss": 0.39116695523262024, | |
| "eval_runtime": 34.2797, | |
| "eval_samples_per_second": 286.321, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 7.313084753203282e-06, | |
| "loss": 0.2623, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.8622516556291391, | |
| "eval_loss": 0.3969860076904297, | |
| "eval_runtime": 34.2798, | |
| "eval_samples_per_second": 286.32, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 7.038713748731034e-06, | |
| "loss": 0.2475, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.862761079979623, | |
| "eval_loss": 0.4156796634197235, | |
| "eval_runtime": 34.2085, | |
| "eval_samples_per_second": 286.917, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.764342744258787e-06, | |
| "loss": 0.2576, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8691798267957208, | |
| "eval_loss": 0.38194480538368225, | |
| "eval_runtime": 34.1983, | |
| "eval_samples_per_second": 287.003, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.490520481795485e-06, | |
| "loss": 0.1544, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.8614365766683647, | |
| "eval_loss": 0.5428063273429871, | |
| "eval_runtime": 34.103, | |
| "eval_samples_per_second": 287.804, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.2161494773232365e-06, | |
| "loss": 0.1431, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_accuracy": 0.8678553234844626, | |
| "eval_loss": 0.509951114654541, | |
| "eval_runtime": 34.213, | |
| "eval_samples_per_second": 286.879, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.94177847285099e-06, | |
| "loss": 0.1481, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.8615384615384616, | |
| "eval_loss": 0.523091733455658, | |
| "eval_runtime": 34.1262, | |
| "eval_samples_per_second": 287.609, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.667407468378742e-06, | |
| "loss": 0.1367, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_accuracy": 0.8649006622516556, | |
| "eval_loss": 0.5161163806915283, | |
| "eval_runtime": 34.1588, | |
| "eval_samples_per_second": 287.335, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.393036463906494e-06, | |
| "loss": 0.1452, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_accuracy": 0.8658176260825268, | |
| "eval_loss": 0.505763053894043, | |
| "eval_runtime": 34.1846, | |
| "eval_samples_per_second": 287.118, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.119214201443192e-06, | |
| "loss": 0.1401, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.8645950076413652, | |
| "eval_loss": 0.5186136960983276, | |
| "eval_runtime": 34.2274, | |
| "eval_samples_per_second": 286.758, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.844843196970945e-06, | |
| "loss": 0.1418, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.8596026490066225, | |
| "eval_loss": 0.5330820083618164, | |
| "eval_runtime": 34.2035, | |
| "eval_samples_per_second": 286.959, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.570472192498697e-06, | |
| "loss": 0.1381, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_accuracy": 0.8654100866021396, | |
| "eval_loss": 0.5446536540985107, | |
| "eval_runtime": 34.125, | |
| "eval_samples_per_second": 287.619, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 4.296101188026449e-06, | |
| "loss": 0.1416, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_accuracy": 0.8622516556291391, | |
| "eval_loss": 0.540454626083374, | |
| "eval_runtime": 34.1047, | |
| "eval_samples_per_second": 287.79, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.0217301835542025e-06, | |
| "loss": 0.1444, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.8650025471217524, | |
| "eval_loss": 0.5085907578468323, | |
| "eval_runtime": 34.0859, | |
| "eval_samples_per_second": 287.949, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.747359179081955e-06, | |
| "loss": 0.1409, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.8621497707590423, | |
| "eval_loss": 0.5341202616691589, | |
| "eval_runtime": 34.1606, | |
| "eval_samples_per_second": 287.319, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 3.473536916618652e-06, | |
| "loss": 0.1348, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_accuracy": 0.8643912379011717, | |
| "eval_loss": 0.5396838784217834, | |
| "eval_runtime": 34.1392, | |
| "eval_samples_per_second": 287.499, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.199165912146405e-06, | |
| "loss": 0.1389, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_accuracy": 0.8635761589403973, | |
| "eval_loss": 0.548418402671814, | |
| "eval_runtime": 34.1565, | |
| "eval_samples_per_second": 287.354, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 2.924794907674157e-06, | |
| "loss": 0.1483, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.86571574121243, | |
| "eval_loss": 0.5054455399513245, | |
| "eval_runtime": 34.1767, | |
| "eval_samples_per_second": 287.184, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.65042390320191e-06, | |
| "loss": 0.1384, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.8688741721854305, | |
| "eval_loss": 0.5029205083847046, | |
| "eval_runtime": 34.1709, | |
| "eval_samples_per_second": 287.233, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.3760528987296626e-06, | |
| "loss": 0.1374, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.8649006622516556, | |
| "eval_loss": 0.5312877893447876, | |
| "eval_runtime": 34.1909, | |
| "eval_samples_per_second": 287.065, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.101681894257415e-06, | |
| "loss": 0.135, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_accuracy": 0.86571574121243, | |
| "eval_loss": 0.5231665372848511, | |
| "eval_runtime": 34.184, | |
| "eval_samples_per_second": 287.123, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 1.8273108897851676e-06, | |
| "loss": 0.1331, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_accuracy": 0.8667345899133979, | |
| "eval_loss": 0.5193073749542236, | |
| "eval_runtime": 34.2139, | |
| "eval_samples_per_second": 286.872, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.5529398853129202e-06, | |
| "loss": 0.1287, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.8639836984207845, | |
| "eval_loss": 0.5389389395713806, | |
| "eval_runtime": 34.2048, | |
| "eval_samples_per_second": 286.948, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.278568880840673e-06, | |
| "loss": 0.1399, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "eval_accuracy": 0.8647987773815589, | |
| "eval_loss": 0.5319520235061646, | |
| "eval_runtime": 34.136, | |
| "eval_samples_per_second": 287.527, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.0041978763684254e-06, | |
| "loss": 0.1371, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.867142129393785, | |
| "eval_loss": 0.5155227184295654, | |
| "eval_runtime": 34.0818, | |
| "eval_samples_per_second": 287.984, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.298268718961781e-07, | |
| "loss": 0.1403, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.8650025471217524, | |
| "eval_loss": 0.5259838104248047, | |
| "eval_runtime": 34.0228, | |
| "eval_samples_per_second": 288.483, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.5655335144181967e-07, | |
| "loss": 0.1402, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.8661232806928171, | |
| "eval_loss": 0.5214368104934692, | |
| "eval_runtime": 34.0674, | |
| "eval_samples_per_second": 288.105, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.8218234696957227e-07, | |
| "loss": 0.1343, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.8656138563423331, | |
| "eval_loss": 0.522437572479248, | |
| "eval_runtime": 34.0148, | |
| "eval_samples_per_second": 288.551, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 36816, | |
| "total_flos": 3.032342720870999e+17, | |
| "train_runtime": 14794.1737, | |
| "train_samples_per_second": 2.489 | |
| } | |
| ], | |
| "max_steps": 36816, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.032342720870999e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |