| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 7.998804113848362, |
| "eval_steps": 500, |
| "global_step": 4176, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.09567089213106912, |
| "grad_norm": 52.15945816040039, |
| "learning_rate": 2.5e-06, |
| "loss": 10.7426, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.19134178426213824, |
| "grad_norm": 38.14910125732422, |
| "learning_rate": 5e-06, |
| "loss": 10.0104, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.2870126763932074, |
| "grad_norm": 39.01994705200195, |
| "learning_rate": 7.5e-06, |
| "loss": 9.6031, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.3826835685242765, |
| "grad_norm": 33.76708984375, |
| "learning_rate": 1e-05, |
| "loss": 9.1839, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.4783544606553456, |
| "grad_norm": 35.43233108520508, |
| "learning_rate": 1.25e-05, |
| "loss": 8.7209, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.5740253527864148, |
| "grad_norm": 29.526994705200195, |
| "learning_rate": 1.5e-05, |
| "loss": 8.2324, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6696962449174838, |
| "grad_norm": 35.25413131713867, |
| "learning_rate": 1.7500000000000002e-05, |
| "loss": 7.7545, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.765367137048553, |
| "grad_norm": 40.43191909790039, |
| "learning_rate": 2e-05, |
| "loss": 7.3315, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.8610380291796221, |
| "grad_norm": 30.023365020751953, |
| "learning_rate": 2.2499999999999998e-05, |
| "loss": 6.963, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.9567089213106912, |
| "grad_norm": 31.878944396972656, |
| "learning_rate": 2.5e-05, |
| "loss": 6.7297, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.9988041138483617, |
| "eval_accuracy": 0.0, |
| "eval_loss": 6.555630207061768, |
| "eval_normalizer": 94012.0, |
| "eval_runtime": 80.2813, |
| "eval_samples_per_second": 665.871, |
| "eval_steps_per_second": 1.308, |
| "step": 522 |
| }, |
| { |
| "epoch": 1.0535756995933987, |
| "grad_norm": 23.939258575439453, |
| "learning_rate": 2.75e-05, |
| "loss": 6.5661, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.1492465917244679, |
| "grad_norm": 28.73314666748047, |
| "learning_rate": 3e-05, |
| "loss": 6.4297, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.244917483855537, |
| "grad_norm": 25.804365158081055, |
| "learning_rate": 3.2500000000000004e-05, |
| "loss": 6.3161, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.3405883759866062, |
| "grad_norm": 28.974842071533203, |
| "learning_rate": 3.5000000000000004e-05, |
| "loss": 6.231, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.4362592681176751, |
| "grad_norm": 30.538211822509766, |
| "learning_rate": 3.75e-05, |
| "loss": 6.1787, |
| "step": 750 |
| }, |
| { |
| "epoch": 1.5319301602487443, |
| "grad_norm": 22.466585159301758, |
| "learning_rate": 4e-05, |
| "loss": 6.1109, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.6276010523798135, |
| "grad_norm": 25.896324157714844, |
| "learning_rate": 4.25e-05, |
| "loss": 6.0128, |
| "step": 850 |
| }, |
| { |
| "epoch": 1.7232719445108826, |
| "grad_norm": 26.904653549194336, |
| "learning_rate": 4.4999999999999996e-05, |
| "loss": 5.9551, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.8189428366419516, |
| "grad_norm": 26.616758346557617, |
| "learning_rate": 4.75e-05, |
| "loss": 5.8732, |
| "step": 950 |
| }, |
| { |
| "epoch": 1.9146137287730207, |
| "grad_norm": 30.21148109436035, |
| "learning_rate": 5e-05, |
| "loss": 5.7962, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.9988041138483617, |
| "eval_accuracy": 0.0, |
| "eval_loss": 5.6822733879089355, |
| "eval_normalizer": 93558.0, |
| "eval_runtime": 78.5931, |
| "eval_samples_per_second": 680.174, |
| "eval_steps_per_second": 1.336, |
| "step": 1044 |
| }, |
| { |
| "epoch": 2.0114805070557282, |
| "grad_norm": 29.994068145751953, |
| "learning_rate": 5.25e-05, |
| "loss": 5.7392, |
| "step": 1050 |
| }, |
| { |
| "epoch": 2.1071513991867974, |
| "grad_norm": 35.065006256103516, |
| "learning_rate": 5.5e-05, |
| "loss": 5.6481, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.2028222913178666, |
| "grad_norm": 37.09373092651367, |
| "learning_rate": 5.75e-05, |
| "loss": 5.5441, |
| "step": 1150 |
| }, |
| { |
| "epoch": 2.2984931834489357, |
| "grad_norm": 32.272159576416016, |
| "learning_rate": 6e-05, |
| "loss": 5.4931, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.394164075580005, |
| "grad_norm": 27.136716842651367, |
| "learning_rate": 6.25e-05, |
| "loss": 5.4324, |
| "step": 1250 |
| }, |
| { |
| "epoch": 2.489834967711074, |
| "grad_norm": 28.377527236938477, |
| "learning_rate": 6.500000000000001e-05, |
| "loss": 5.3281, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.5855058598421428, |
| "grad_norm": 25.35126495361328, |
| "learning_rate": 6.75e-05, |
| "loss": 5.2737, |
| "step": 1350 |
| }, |
| { |
| "epoch": 2.6811767519732124, |
| "grad_norm": 27.91370964050293, |
| "learning_rate": 7.000000000000001e-05, |
| "loss": 5.1901, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.776847644104281, |
| "grad_norm": 24.87859344482422, |
| "learning_rate": 7.25e-05, |
| "loss": 5.141, |
| "step": 1450 |
| }, |
| { |
| "epoch": 2.8725185362353503, |
| "grad_norm": 27.901790618896484, |
| "learning_rate": 7.5e-05, |
| "loss": 5.0599, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.9681894283664194, |
| "grad_norm": 25.183063507080078, |
| "learning_rate": 7.75e-05, |
| "loss": 5.034, |
| "step": 1550 |
| }, |
| { |
| "epoch": 2.9988041138483617, |
| "eval_accuracy": 0.0, |
| "eval_loss": 4.937503337860107, |
| "eval_normalizer": 94216.0, |
| "eval_runtime": 75.5016, |
| "eval_samples_per_second": 708.024, |
| "eval_steps_per_second": 1.391, |
| "step": 1566 |
| }, |
| { |
| "epoch": 3.065056206649127, |
| "grad_norm": 26.59418487548828, |
| "learning_rate": 8e-05, |
| "loss": 4.9628, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.160727098780196, |
| "grad_norm": 25.259281158447266, |
| "learning_rate": 8.25e-05, |
| "loss": 4.9004, |
| "step": 1650 |
| }, |
| { |
| "epoch": 3.2563979909112652, |
| "grad_norm": 25.84400749206543, |
| "learning_rate": 8.5e-05, |
| "loss": 4.8452, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.3520688830423344, |
| "grad_norm": 25.52320671081543, |
| "learning_rate": 8.75e-05, |
| "loss": 4.7968, |
| "step": 1750 |
| }, |
| { |
| "epoch": 3.4477397751734036, |
| "grad_norm": 26.29749298095703, |
| "learning_rate": 8.999999999999999e-05, |
| "loss": 4.7601, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.5434106673044727, |
| "grad_norm": 27.535959243774414, |
| "learning_rate": 9.25e-05, |
| "loss": 4.7155, |
| "step": 1850 |
| }, |
| { |
| "epoch": 3.639081559435542, |
| "grad_norm": 23.676984786987305, |
| "learning_rate": 9.5e-05, |
| "loss": 4.6651, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.734752451566611, |
| "grad_norm": 22.59883689880371, |
| "learning_rate": 9.750000000000001e-05, |
| "loss": 4.6527, |
| "step": 1950 |
| }, |
| { |
| "epoch": 3.83042334369768, |
| "grad_norm": 21.733461380004883, |
| "learning_rate": 0.0001, |
| "loss": 4.616, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.926094235828749, |
| "grad_norm": 25.904829025268555, |
| "learning_rate": 0.0001025, |
| "loss": 4.5769, |
| "step": 2050 |
| }, |
| { |
| "epoch": 3.9988041138483617, |
| "eval_accuracy": 0.0, |
| "eval_loss": 4.510604381561279, |
| "eval_normalizer": 93665.0, |
| "eval_runtime": 76.2941, |
| "eval_samples_per_second": 700.67, |
| "eval_steps_per_second": 1.376, |
| "step": 2088 |
| }, |
| { |
| "epoch": 4.0229610141114565, |
| "grad_norm": 21.005359649658203, |
| "learning_rate": 0.000105, |
| "loss": 4.5525, |
| "step": 2100 |
| }, |
| { |
| "epoch": 4.118631906242526, |
| "grad_norm": 25.174129486083984, |
| "learning_rate": 0.0001075, |
| "loss": 4.4938, |
| "step": 2150 |
| }, |
| { |
| "epoch": 4.214302798373595, |
| "grad_norm": 22.09128189086914, |
| "learning_rate": 0.00011, |
| "loss": 4.4503, |
| "step": 2200 |
| }, |
| { |
| "epoch": 4.309973690504664, |
| "grad_norm": 21.93778419494629, |
| "learning_rate": 0.00011250000000000001, |
| "loss": 4.4222, |
| "step": 2250 |
| }, |
| { |
| "epoch": 4.405644582635733, |
| "grad_norm": 22.780986785888672, |
| "learning_rate": 0.000115, |
| "loss": 4.388, |
| "step": 2300 |
| }, |
| { |
| "epoch": 4.501315474766802, |
| "grad_norm": 24.103681564331055, |
| "learning_rate": 0.0001175, |
| "loss": 4.3805, |
| "step": 2350 |
| }, |
| { |
| "epoch": 4.596986366897871, |
| "grad_norm": 27.335912704467773, |
| "learning_rate": 0.00012, |
| "loss": 4.3188, |
| "step": 2400 |
| }, |
| { |
| "epoch": 4.69265725902894, |
| "grad_norm": 21.6763916015625, |
| "learning_rate": 0.0001225, |
| "loss": 4.2918, |
| "step": 2450 |
| }, |
| { |
| "epoch": 4.78832815116001, |
| "grad_norm": 24.06027603149414, |
| "learning_rate": 0.000125, |
| "loss": 4.2723, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.8839990432910785, |
| "grad_norm": 23.394180297851562, |
| "learning_rate": 0.0001275, |
| "loss": 4.2586, |
| "step": 2550 |
| }, |
| { |
| "epoch": 4.979669935422148, |
| "grad_norm": 25.303680419921875, |
| "learning_rate": 0.00013000000000000002, |
| "loss": 4.2298, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.998804113848362, |
| "eval_accuracy": 0.0, |
| "eval_loss": 4.226888656616211, |
| "eval_normalizer": 94313.0, |
| "eval_runtime": 77.1025, |
| "eval_samples_per_second": 693.324, |
| "eval_steps_per_second": 1.362, |
| "step": 2610 |
| }, |
| { |
| "epoch": 5.076536713704855, |
| "grad_norm": 23.319316864013672, |
| "learning_rate": 0.00013250000000000002, |
| "loss": 4.2049, |
| "step": 2650 |
| }, |
| { |
| "epoch": 5.172207605835925, |
| "grad_norm": 27.484468460083008, |
| "learning_rate": 0.000135, |
| "loss": 4.1758, |
| "step": 2700 |
| }, |
| { |
| "epoch": 5.2678784979669935, |
| "grad_norm": 25.021984100341797, |
| "learning_rate": 0.0001375, |
| "loss": 4.1389, |
| "step": 2750 |
| }, |
| { |
| "epoch": 5.363549390098063, |
| "grad_norm": 24.843420028686523, |
| "learning_rate": 0.00014000000000000001, |
| "loss": 4.123, |
| "step": 2800 |
| }, |
| { |
| "epoch": 5.459220282229132, |
| "grad_norm": 22.530752182006836, |
| "learning_rate": 0.0001425, |
| "loss": 4.1117, |
| "step": 2850 |
| }, |
| { |
| "epoch": 5.5548911743602005, |
| "grad_norm": 23.811887741088867, |
| "learning_rate": 0.000145, |
| "loss": 4.0883, |
| "step": 2900 |
| }, |
| { |
| "epoch": 5.65056206649127, |
| "grad_norm": 22.435691833496094, |
| "learning_rate": 0.0001475, |
| "loss": 4.0598, |
| "step": 2950 |
| }, |
| { |
| "epoch": 5.746232958622339, |
| "grad_norm": 22.16675567626953, |
| "learning_rate": 0.00015, |
| "loss": 4.0562, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.8419038507534085, |
| "grad_norm": 26.84369468688965, |
| "learning_rate": 0.0001525, |
| "loss": 4.0424, |
| "step": 3050 |
| }, |
| { |
| "epoch": 5.937574742884477, |
| "grad_norm": 22.534196853637695, |
| "learning_rate": 0.000155, |
| "loss": 4.0289, |
| "step": 3100 |
| }, |
| { |
| "epoch": 5.998804113848362, |
| "eval_accuracy": 0.0, |
| "eval_loss": 4.000634670257568, |
| "eval_normalizer": 93901.0, |
| "eval_runtime": 76.1749, |
| "eval_samples_per_second": 701.767, |
| "eval_steps_per_second": 1.378, |
| "step": 3132 |
| }, |
| { |
| "epoch": 6.034441521167185, |
| "grad_norm": 22.28015899658203, |
| "learning_rate": 0.0001575, |
| "loss": 4.0049, |
| "step": 3150 |
| }, |
| { |
| "epoch": 6.130112413298254, |
| "grad_norm": 22.31985855102539, |
| "learning_rate": 0.00016, |
| "loss": 3.9543, |
| "step": 3200 |
| }, |
| { |
| "epoch": 6.2257833054293235, |
| "grad_norm": 26.62820053100586, |
| "learning_rate": 0.00016250000000000002, |
| "loss": 3.96, |
| "step": 3250 |
| }, |
| { |
| "epoch": 6.321454197560392, |
| "grad_norm": 22.462852478027344, |
| "learning_rate": 0.000165, |
| "loss": 3.9324, |
| "step": 3300 |
| }, |
| { |
| "epoch": 6.417125089691462, |
| "grad_norm": 21.36258888244629, |
| "learning_rate": 0.0001675, |
| "loss": 3.9276, |
| "step": 3350 |
| }, |
| { |
| "epoch": 6.5127959818225305, |
| "grad_norm": 21.731992721557617, |
| "learning_rate": 0.00017, |
| "loss": 3.8995, |
| "step": 3400 |
| }, |
| { |
| "epoch": 6.608466873953599, |
| "grad_norm": 21.62281608581543, |
| "learning_rate": 0.0001725, |
| "loss": 3.9031, |
| "step": 3450 |
| }, |
| { |
| "epoch": 6.704137766084669, |
| "grad_norm": 22.04605484008789, |
| "learning_rate": 0.000175, |
| "loss": 3.8636, |
| "step": 3500 |
| }, |
| { |
| "epoch": 6.7998086582157375, |
| "grad_norm": 22.352691650390625, |
| "learning_rate": 0.0001775, |
| "loss": 3.8825, |
| "step": 3550 |
| }, |
| { |
| "epoch": 6.895479550346807, |
| "grad_norm": 23.3765811920166, |
| "learning_rate": 0.00017999999999999998, |
| "loss": 3.842, |
| "step": 3600 |
| }, |
| { |
| "epoch": 6.991150442477876, |
| "grad_norm": 21.487628936767578, |
| "learning_rate": 0.0001825, |
| "loss": 3.8673, |
| "step": 3650 |
| }, |
| { |
| "epoch": 6.998804113848362, |
| "eval_accuracy": 0.0, |
| "eval_loss": 3.867459774017334, |
| "eval_normalizer": 94462.0, |
| "eval_runtime": 77.3784, |
| "eval_samples_per_second": 690.852, |
| "eval_steps_per_second": 1.357, |
| "step": 3654 |
| }, |
| { |
| "epoch": 7.088017220760584, |
| "grad_norm": 21.705764770507812, |
| "learning_rate": 0.000185, |
| "loss": 3.794, |
| "step": 3700 |
| }, |
| { |
| "epoch": 7.1836881128916525, |
| "grad_norm": 21.813831329345703, |
| "learning_rate": 0.0001875, |
| "loss": 3.8044, |
| "step": 3750 |
| }, |
| { |
| "epoch": 7.279359005022722, |
| "grad_norm": 23.83123779296875, |
| "learning_rate": 0.00019, |
| "loss": 3.7978, |
| "step": 3800 |
| }, |
| { |
| "epoch": 7.375029897153791, |
| "grad_norm": 22.62957000732422, |
| "learning_rate": 0.00019250000000000002, |
| "loss": 3.7845, |
| "step": 3850 |
| }, |
| { |
| "epoch": 7.4707007892848605, |
| "grad_norm": 23.31735610961914, |
| "learning_rate": 0.00019500000000000002, |
| "loss": 3.7775, |
| "step": 3900 |
| }, |
| { |
| "epoch": 7.566371681415929, |
| "grad_norm": 21.118228912353516, |
| "learning_rate": 0.0001975, |
| "loss": 3.7552, |
| "step": 3950 |
| }, |
| { |
| "epoch": 7.662042573546998, |
| "grad_norm": 21.326833724975586, |
| "learning_rate": 0.0002, |
| "loss": 3.7476, |
| "step": 4000 |
| }, |
| { |
| "epoch": 7.7577134656780675, |
| "grad_norm": 23.456077575683594, |
| "learning_rate": 0.00020250000000000002, |
| "loss": 3.7478, |
| "step": 4050 |
| }, |
| { |
| "epoch": 7.853384357809136, |
| "grad_norm": 23.339223861694336, |
| "learning_rate": 0.000205, |
| "loss": 3.7394, |
| "step": 4100 |
| }, |
| { |
| "epoch": 7.949055249940206, |
| "grad_norm": 22.789596557617188, |
| "learning_rate": 0.0002075, |
| "loss": 3.7111, |
| "step": 4150 |
| }, |
| { |
| "epoch": 7.998804113848362, |
| "eval_accuracy": 0.0, |
| "eval_loss": 3.691098928451538, |
| "eval_normalizer": 93833.0, |
| "eval_runtime": 76.6695, |
| "eval_samples_per_second": 697.239, |
| "eval_steps_per_second": 1.37, |
| "step": 4176 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 5220, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "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": 6.61778132585005e+17, |
| "train_batch_size": 128, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
