| { | |
| "best_metric": 2.206617832183838, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.0768344218209758, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001536688436419516, | |
| "eval_loss": 3.802302360534668, | |
| "eval_runtime": 176.2084, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 3.887, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001536688436419516, | |
| "grad_norm": 29.49568748474121, | |
| "learning_rate": 4.16e-05, | |
| "loss": 5.6992, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.003073376872839032, | |
| "grad_norm": 11.658048629760742, | |
| "learning_rate": 8.32e-05, | |
| "loss": 6.3089, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004610065309258548, | |
| "grad_norm": 14.371132850646973, | |
| "learning_rate": 0.0001248, | |
| "loss": 5.7372, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.006146753745678064, | |
| "grad_norm": 18.474430084228516, | |
| "learning_rate": 0.0001664, | |
| "loss": 6.365, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00768344218209758, | |
| "grad_norm": 42.71783447265625, | |
| "learning_rate": 0.000208, | |
| "loss": 7.0134, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00768344218209758, | |
| "eval_loss": 3.467595338821411, | |
| "eval_runtime": 175.6638, | |
| "eval_samples_per_second": 15.598, | |
| "eval_steps_per_second": 3.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.009220130618517095, | |
| "grad_norm": 8.05245590209961, | |
| "learning_rate": 0.0002077466612270217, | |
| "loss": 5.4029, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.010756819054936612, | |
| "grad_norm": 7.458392143249512, | |
| "learning_rate": 0.0002069878791491233, | |
| "loss": 5.7947, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.012293507491356128, | |
| "grad_norm": 27.035818099975586, | |
| "learning_rate": 0.00020572735047631578, | |
| "loss": 5.5678, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.013830195927775643, | |
| "grad_norm": 17.315340042114258, | |
| "learning_rate": 0.00020397121637758515, | |
| "loss": 6.5471, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.01536688436419516, | |
| "grad_norm": 33.20652770996094, | |
| "learning_rate": 0.00020172803256173445, | |
| "loss": 7.03, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01536688436419516, | |
| "eval_loss": 3.3676817417144775, | |
| "eval_runtime": 176.2178, | |
| "eval_samples_per_second": 15.549, | |
| "eval_steps_per_second": 3.887, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.016903572800614674, | |
| "grad_norm": 6.50840425491333, | |
| "learning_rate": 0.00019900872759483047, | |
| "loss": 5.6033, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.01844026123703419, | |
| "grad_norm": 6.7641825675964355, | |
| "learning_rate": 0.0001958265496573284, | |
| "loss": 5.4601, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.019976949673453707, | |
| "grad_norm": 14.109837532043457, | |
| "learning_rate": 0.00019219700200026827, | |
| "loss": 5.4273, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.021513638109873223, | |
| "grad_norm": 11.948671340942383, | |
| "learning_rate": 0.0001881377674149945, | |
| "loss": 6.1093, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.02305032654629274, | |
| "grad_norm": 21.177404403686523, | |
| "learning_rate": 0.00018366862208437368, | |
| "loss": 6.2444, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.02305032654629274, | |
| "eval_loss": 3.1878161430358887, | |
| "eval_runtime": 175.8544, | |
| "eval_samples_per_second": 15.581, | |
| "eval_steps_per_second": 3.895, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.024587014982712256, | |
| "grad_norm": 6.276320457458496, | |
| "learning_rate": 0.00017881133923521971, | |
| "loss": 5.5092, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.026123703419131773, | |
| "grad_norm": 6.799064636230469, | |
| "learning_rate": 0.00017358958306132124, | |
| "loss": 5.1537, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.027660391855551286, | |
| "grad_norm": 9.907474517822266, | |
| "learning_rate": 0.00016802879343386844, | |
| "loss": 5.5056, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.029197080291970802, | |
| "grad_norm": 13.780465126037598, | |
| "learning_rate": 0.00016215606196095766, | |
| "loss": 5.6722, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.03073376872839032, | |
| "grad_norm": 17.128713607788086, | |
| "learning_rate": 0.000156, | |
| "loss": 5.9664, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.03073376872839032, | |
| "eval_loss": 3.0006656646728516, | |
| "eval_runtime": 175.6129, | |
| "eval_samples_per_second": 15.602, | |
| "eval_steps_per_second": 3.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.032270457164809835, | |
| "grad_norm": 4.835846900939941, | |
| "learning_rate": 0.00014959059926606403, | |
| "loss": 5.1, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.03380714560122935, | |
| "grad_norm": 5.523816108703613, | |
| "learning_rate": 0.00014295908571525487, | |
| "loss": 4.9443, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.03534383403764887, | |
| "grad_norm": 9.212928771972656, | |
| "learning_rate": 0.00013613776741499452, | |
| "loss": 4.8833, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.03688052247406838, | |
| "grad_norm": 10.690314292907715, | |
| "learning_rate": 0.00012915987714236542, | |
| "loss": 5.425, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.0384172109104879, | |
| "grad_norm": 16.973690032958984, | |
| "learning_rate": 0.00012205941047736077, | |
| "loss": 5.6566, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0384172109104879, | |
| "eval_loss": 2.7571074962615967, | |
| "eval_runtime": 175.489, | |
| "eval_samples_per_second": 15.614, | |
| "eval_steps_per_second": 3.903, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.039953899346907414, | |
| "grad_norm": 4.981348514556885, | |
| "learning_rate": 0.00011487096017983597, | |
| "loss": 4.7851, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.041490587783326933, | |
| "grad_norm": 6.857975482940674, | |
| "learning_rate": 0.00010762954765706012, | |
| "loss": 4.6581, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.043027276219746446, | |
| "grad_norm": 8.468694686889648, | |
| "learning_rate": 0.00010037045234293992, | |
| "loss": 4.5606, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.04456396465616596, | |
| "grad_norm": 11.279480934143066, | |
| "learning_rate": 9.312903982016405e-05, | |
| "loss": 5.3172, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.04610065309258548, | |
| "grad_norm": 19.332422256469727, | |
| "learning_rate": 8.594058952263925e-05, | |
| "loss": 5.3847, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.04610065309258548, | |
| "eval_loss": 2.572946310043335, | |
| "eval_runtime": 175.6968, | |
| "eval_samples_per_second": 15.595, | |
| "eval_steps_per_second": 3.899, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.04763734152900499, | |
| "grad_norm": 4.437800407409668, | |
| "learning_rate": 7.884012285763457e-05, | |
| "loss": 4.644, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.04917402996542451, | |
| "grad_norm": 4.734920024871826, | |
| "learning_rate": 7.186223258500548e-05, | |
| "loss": 4.554, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.050710718401844025, | |
| "grad_norm": 7.473479747772217, | |
| "learning_rate": 6.504091428474514e-05, | |
| "loss": 4.4863, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.052247406838263545, | |
| "grad_norm": 8.95118236541748, | |
| "learning_rate": 5.840940073393593e-05, | |
| "loss": 5.0514, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.05378409527468306, | |
| "grad_norm": 15.974135398864746, | |
| "learning_rate": 5.200000000000002e-05, | |
| "loss": 5.0546, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.05378409527468306, | |
| "eval_loss": 2.400266170501709, | |
| "eval_runtime": 178.4074, | |
| "eval_samples_per_second": 15.358, | |
| "eval_steps_per_second": 3.84, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.05532078371110257, | |
| "grad_norm": 3.9929511547088623, | |
| "learning_rate": 4.5843938039042344e-05, | |
| "loss": 4.4327, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.05685747214752209, | |
| "grad_norm": 5.959095478057861, | |
| "learning_rate": 3.997120656613154e-05, | |
| "loss": 4.501, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.058394160583941604, | |
| "grad_norm": 7.357118606567383, | |
| "learning_rate": 3.441041693867878e-05, | |
| "loss": 4.2774, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.059930849020361124, | |
| "grad_norm": 10.237184524536133, | |
| "learning_rate": 2.9188660764780296e-05, | |
| "loss": 4.8857, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.06146753745678064, | |
| "grad_norm": 14.245146751403809, | |
| "learning_rate": 2.4331377915626298e-05, | |
| "loss": 4.9947, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06146753745678064, | |
| "eval_loss": 2.4386484622955322, | |
| "eval_runtime": 175.8523, | |
| "eval_samples_per_second": 15.581, | |
| "eval_steps_per_second": 3.895, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06300422589320015, | |
| "grad_norm": 3.9970903396606445, | |
| "learning_rate": 1.9862232585005475e-05, | |
| "loss": 4.2686, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.06454091432961967, | |
| "grad_norm": 6.5729546546936035, | |
| "learning_rate": 1.58029979997317e-05, | |
| "loss": 4.3436, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.06607760276603919, | |
| "grad_norm": 7.991576194763184, | |
| "learning_rate": 1.2173450342671593e-05, | |
| "loss": 3.9325, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.0676142912024587, | |
| "grad_norm": 11.29770278930664, | |
| "learning_rate": 8.991272405169498e-06, | |
| "loss": 4.787, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.06915097963887822, | |
| "grad_norm": 15.63181209564209, | |
| "learning_rate": 6.271967438265535e-06, | |
| "loss": 5.0258, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.06915097963887822, | |
| "eval_loss": 2.2233057022094727, | |
| "eval_runtime": 176.0184, | |
| "eval_samples_per_second": 15.567, | |
| "eval_steps_per_second": 3.892, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.07068766807529774, | |
| "grad_norm": 4.114557266235352, | |
| "learning_rate": 4.028783622414835e-06, | |
| "loss": 4.1326, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.07222435651171726, | |
| "grad_norm": 4.679042339324951, | |
| "learning_rate": 2.272649523684208e-06, | |
| "loss": 4.2644, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.07376104494813676, | |
| "grad_norm": 6.639362335205078, | |
| "learning_rate": 1.0121208508766823e-06, | |
| "loss": 4.0476, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.07529773338455628, | |
| "grad_norm": 8.920401573181152, | |
| "learning_rate": 2.533387729782834e-07, | |
| "loss": 4.6659, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.0768344218209758, | |
| "grad_norm": 13.285958290100098, | |
| "learning_rate": 0.0, | |
| "loss": 5.0228, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0768344218209758, | |
| "eval_loss": 2.206617832183838, | |
| "eval_runtime": 175.6859, | |
| "eval_samples_per_second": 15.596, | |
| "eval_steps_per_second": 3.899, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.829167574233907e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |