| { | |
| "best_metric": 0.8063241106719368, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-pi33ffzs/checkpoint-800", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 800, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 14.099751472473145, | |
| "learning_rate": 9.582405619536138e-05, | |
| "loss": 0.6743, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 4.672331809997559, | |
| "learning_rate": 9.522138917526476e-05, | |
| "loss": 0.6209, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 2.4487009048461914, | |
| "learning_rate": 9.461872215516814e-05, | |
| "loss": 0.608, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.88143253326416, | |
| "learning_rate": 9.401605513507153e-05, | |
| "loss": 0.5719, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 7.989026069641113, | |
| "learning_rate": 9.341338811497491e-05, | |
| "loss": 0.5645, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.12394380569458, | |
| "learning_rate": 9.281072109487831e-05, | |
| "loss": 0.4137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 6.6350579261779785, | |
| "learning_rate": 9.22080540747817e-05, | |
| "loss": 0.5479, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.6092991828918457, | |
| "learning_rate": 9.160538705468508e-05, | |
| "loss": 0.5195, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 22.4118595123291, | |
| "learning_rate": 9.100272003458847e-05, | |
| "loss": 0.8602, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.869532585144043, | |
| "learning_rate": 9.040005301449186e-05, | |
| "loss": 0.9557, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 5.411820888519287, | |
| "learning_rate": 8.979738599439524e-05, | |
| "loss": 0.4593, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.3322949409484863, | |
| "learning_rate": 8.919471897429864e-05, | |
| "loss": 0.5916, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 2.0190422534942627, | |
| "learning_rate": 8.859205195420202e-05, | |
| "loss": 0.5737, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 6.8667893409729, | |
| "learning_rate": 8.798938493410541e-05, | |
| "loss": 0.4983, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 1.2699512243270874, | |
| "learning_rate": 8.738671791400879e-05, | |
| "loss": 0.5618, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.28828763961792, | |
| "learning_rate": 8.678405089391219e-05, | |
| "loss": 0.7923, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 9.879742622375488, | |
| "learning_rate": 8.618138387381557e-05, | |
| "loss": 0.8174, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.6045369505882263, | |
| "learning_rate": 8.557871685371895e-05, | |
| "loss": 0.284, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 0.8522588610649109, | |
| "learning_rate": 8.497604983362234e-05, | |
| "loss": 0.3718, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 9.719417572021484, | |
| "learning_rate": 8.437338281352574e-05, | |
| "loss": 1.0217, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 15.228802680969238, | |
| "learning_rate": 8.377071579342912e-05, | |
| "loss": 0.5518, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 6.107767105102539, | |
| "learning_rate": 8.316804877333251e-05, | |
| "loss": 0.5761, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 3.676274061203003, | |
| "learning_rate": 8.25653817532359e-05, | |
| "loss": 0.2779, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.597080647945404, | |
| "learning_rate": 8.196271473313928e-05, | |
| "loss": 0.2176, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.3770584762096405, | |
| "learning_rate": 8.136004771304267e-05, | |
| "loss": 0.6285, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 23.122947692871094, | |
| "learning_rate": 8.075738069294607e-05, | |
| "loss": 0.5774, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 0.3317395746707916, | |
| "learning_rate": 8.015471367284945e-05, | |
| "loss": 0.76, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.390415906906128, | |
| "learning_rate": 7.955204665275283e-05, | |
| "loss": 0.502, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 4.841302871704102, | |
| "learning_rate": 7.894937963265622e-05, | |
| "loss": 0.4339, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 7.026674747467041, | |
| "learning_rate": 7.83467126125596e-05, | |
| "loss": 0.6468, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 0.8100307583808899, | |
| "learning_rate": 7.7744045592463e-05, | |
| "loss": 0.6812, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.845278739929199, | |
| "learning_rate": 7.714137857236639e-05, | |
| "loss": 0.3495, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 22.603090286254883, | |
| "learning_rate": 7.653871155226977e-05, | |
| "loss": 0.5564, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.4318823218345642, | |
| "learning_rate": 7.593604453217316e-05, | |
| "loss": 0.5139, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 12.046862602233887, | |
| "learning_rate": 7.533337751207655e-05, | |
| "loss": 0.7725, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 15.319775581359863, | |
| "learning_rate": 7.473071049197993e-05, | |
| "loss": 0.4356, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 25.348968505859375, | |
| "learning_rate": 7.412804347188333e-05, | |
| "loss": 0.4134, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 21.941987991333008, | |
| "learning_rate": 7.35253764517867e-05, | |
| "loss": 0.8063, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 4.438033103942871, | |
| "learning_rate": 7.29227094316901e-05, | |
| "loss": 0.5669, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.872718811035156, | |
| "learning_rate": 7.232004241159348e-05, | |
| "loss": 0.5456, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8075, | |
| "eval_f1": 0.5792349726775956, | |
| "eval_loss": 0.5038720369338989, | |
| "eval_precision": 0.8833333333333333, | |
| "eval_recall": 0.43089430894308944, | |
| "eval_runtime": 1.5184, | |
| "eval_samples_per_second": 263.438, | |
| "eval_steps_per_second": 16.465, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.025, | |
| "grad_norm": 0.6440209150314331, | |
| "learning_rate": 7.171737539149688e-05, | |
| "loss": 0.3559, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 10.962855339050293, | |
| "learning_rate": 7.111470837140027e-05, | |
| "loss": 0.2384, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.075, | |
| "grad_norm": 0.5640668869018555, | |
| "learning_rate": 7.051204135130364e-05, | |
| "loss": 0.5156, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.4815268814563751, | |
| "learning_rate": 6.990937433120703e-05, | |
| "loss": 0.2444, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 6.384509086608887, | |
| "learning_rate": 6.930670731111043e-05, | |
| "loss": 0.2974, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.1514960080385208, | |
| "learning_rate": 6.870404029101381e-05, | |
| "loss": 0.3388, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.175, | |
| "grad_norm": 8.39075756072998, | |
| "learning_rate": 6.81013732709172e-05, | |
| "loss": 0.7306, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.8284034132957458, | |
| "learning_rate": 6.749870625082059e-05, | |
| "loss": 0.5125, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.225, | |
| "grad_norm": 7.400186538696289, | |
| "learning_rate": 6.689603923072397e-05, | |
| "loss": 0.2146, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.23411034047603607, | |
| "learning_rate": 6.629337221062736e-05, | |
| "loss": 0.2123, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.275, | |
| "grad_norm": 4.417282581329346, | |
| "learning_rate": 6.569070519053076e-05, | |
| "loss": 0.5131, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 5.702905178070068, | |
| "learning_rate": 6.508803817043414e-05, | |
| "loss": 0.794, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.325, | |
| "grad_norm": 0.25387826561927795, | |
| "learning_rate": 6.448537115033752e-05, | |
| "loss": 0.268, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 4.115130424499512, | |
| "learning_rate": 6.388270413024091e-05, | |
| "loss": 0.5022, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 1.3796603679656982, | |
| "learning_rate": 6.32800371101443e-05, | |
| "loss": 0.4664, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 9.150830268859863, | |
| "learning_rate": 6.267737009004769e-05, | |
| "loss": 0.496, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.425, | |
| "grad_norm": 0.4542056918144226, | |
| "learning_rate": 6.207470306995108e-05, | |
| "loss": 0.2549, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 29.90502166748047, | |
| "learning_rate": 6.147203604985446e-05, | |
| "loss": 0.6464, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.475, | |
| "grad_norm": 1.0268361568450928, | |
| "learning_rate": 6.0869369029757845e-05, | |
| "loss": 0.2584, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.35682424902915955, | |
| "learning_rate": 6.026670200966124e-05, | |
| "loss": 0.5534, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.525, | |
| "grad_norm": 8.86542797088623, | |
| "learning_rate": 5.966403498956463e-05, | |
| "loss": 0.5656, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 8.581369400024414, | |
| "learning_rate": 5.9061367969468016e-05, | |
| "loss": 0.5376, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.575, | |
| "grad_norm": 4.579167366027832, | |
| "learning_rate": 5.84587009493714e-05, | |
| "loss": 0.3394, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.4072767198085785, | |
| "learning_rate": 5.7856033929274785e-05, | |
| "loss": 0.8387, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.8675380945205688, | |
| "learning_rate": 5.725336690917818e-05, | |
| "loss": 0.4963, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 5.402355670928955, | |
| "learning_rate": 5.665069988908157e-05, | |
| "loss": 0.6346, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.675, | |
| "grad_norm": 0.9370933175086975, | |
| "learning_rate": 5.6048032868984955e-05, | |
| "loss": 0.3953, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 5.052032947540283, | |
| "learning_rate": 5.5445365848888336e-05, | |
| "loss": 0.4608, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.725, | |
| "grad_norm": 0.40091681480407715, | |
| "learning_rate": 5.4842698828791724e-05, | |
| "loss": 0.4279, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.348732590675354, | |
| "learning_rate": 5.424003180869511e-05, | |
| "loss": 0.0545, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.775, | |
| "grad_norm": 0.14755772054195404, | |
| "learning_rate": 5.3637364788598506e-05, | |
| "loss": 0.5262, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 11.312381744384766, | |
| "learning_rate": 5.3034697768501894e-05, | |
| "loss": 0.5846, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.825, | |
| "grad_norm": 0.31017163395881653, | |
| "learning_rate": 5.2432030748405275e-05, | |
| "loss": 0.4936, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.26326823234558105, | |
| "learning_rate": 5.182936372830866e-05, | |
| "loss": 0.1889, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.2871868312358856, | |
| "learning_rate": 5.122669670821205e-05, | |
| "loss": 0.1512, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.174193874001503, | |
| "learning_rate": 5.062402968811544e-05, | |
| "loss": 0.1859, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.925, | |
| "grad_norm": 5.3941330909729, | |
| "learning_rate": 5.0021362668018834e-05, | |
| "loss": 0.8953, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 5.847171783447266, | |
| "learning_rate": 4.941869564792221e-05, | |
| "loss": 0.4321, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.975, | |
| "grad_norm": 5.563133716583252, | |
| "learning_rate": 4.88160286278256e-05, | |
| "loss": 0.61, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.7091472148895264, | |
| "learning_rate": 4.821336160772899e-05, | |
| "loss": 0.3231, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8775, | |
| "eval_f1": 0.8063241106719368, | |
| "eval_loss": 0.375855952501297, | |
| "eval_precision": 0.7846153846153846, | |
| "eval_recall": 0.8292682926829268, | |
| "eval_runtime": 1.5214, | |
| "eval_samples_per_second": 262.914, | |
| "eval_steps_per_second": 16.432, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423630740901888.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 9.642672321545798e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 4, | |
| "seed": 12 | |
| } | |
| } | |