| { | |
| "best_metric": 0.4194607138633728, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.37091988130563797, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000741839762611276, | |
| "eval_loss": 0.6393800377845764, | |
| "eval_runtime": 63.0989, | |
| "eval_samples_per_second": 9.002, | |
| "eval_steps_per_second": 2.25, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00741839762611276, | |
| "grad_norm": 0.4622786045074463, | |
| "learning_rate": 4.24e-05, | |
| "loss": 0.641, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01483679525222552, | |
| "grad_norm": 0.6135218739509583, | |
| "learning_rate": 8.48e-05, | |
| "loss": 0.5886, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02225519287833828, | |
| "grad_norm": 0.36783531308174133, | |
| "learning_rate": 0.0001272, | |
| "loss": 0.5411, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02967359050445104, | |
| "grad_norm": 0.40542206168174744, | |
| "learning_rate": 0.0001696, | |
| "loss": 0.4827, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.037091988130563795, | |
| "grad_norm": 0.4334207773208618, | |
| "learning_rate": 0.000212, | |
| "loss": 0.463, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.037091988130563795, | |
| "eval_loss": 0.4923464357852936, | |
| "eval_runtime": 63.1469, | |
| "eval_samples_per_second": 8.995, | |
| "eval_steps_per_second": 2.249, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04451038575667656, | |
| "grad_norm": 0.4054912030696869, | |
| "learning_rate": 0.00021174178932754136, | |
| "loss": 0.5025, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05192878338278932, | |
| "grad_norm": 0.39622294902801514, | |
| "learning_rate": 0.00021096841528660647, | |
| "loss": 0.504, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05934718100890208, | |
| "grad_norm": 0.4143044054508209, | |
| "learning_rate": 0.0002096836456777834, | |
| "loss": 0.4841, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06676557863501484, | |
| "grad_norm": 0.39025792479515076, | |
| "learning_rate": 0.00020789373976946182, | |
| "loss": 0.4727, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07418397626112759, | |
| "grad_norm": 0.3781468868255615, | |
| "learning_rate": 0.0002056074178033063, | |
| "loss": 0.4287, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07418397626112759, | |
| "eval_loss": 0.47025901079177856, | |
| "eval_runtime": 63.1627, | |
| "eval_samples_per_second": 8.993, | |
| "eval_steps_per_second": 2.248, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08160237388724036, | |
| "grad_norm": 0.43171194195747375, | |
| "learning_rate": 0.00020283581851011567, | |
| "loss": 0.4821, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08902077151335312, | |
| "grad_norm": 0.3957735002040863, | |
| "learning_rate": 0.00019959244484304625, | |
| "loss": 0.5191, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09643916913946587, | |
| "grad_norm": 0.40845993161201477, | |
| "learning_rate": 0.00019589309819258114, | |
| "loss": 0.4756, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.10385756676557864, | |
| "grad_norm": 0.42555275559425354, | |
| "learning_rate": 0.00019175580140374444, | |
| "loss": 0.4571, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.11127596439169139, | |
| "grad_norm": 0.39264148473739624, | |
| "learning_rate": 0.00018720071097061167, | |
| "loss": 0.4268, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11127596439169139, | |
| "eval_loss": 0.45653077960014343, | |
| "eval_runtime": 63.1089, | |
| "eval_samples_per_second": 9.0, | |
| "eval_steps_per_second": 2.25, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11869436201780416, | |
| "grad_norm": 0.4264899790287018, | |
| "learning_rate": 0.00018225001883589702, | |
| "loss": 0.4657, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1261127596439169, | |
| "grad_norm": 0.39543238282203674, | |
| "learning_rate": 0.00017692784427403898, | |
| "loss": 0.5229, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.13353115727002968, | |
| "grad_norm": 0.38356471061706543, | |
| "learning_rate": 0.00017126011638451976, | |
| "loss": 0.4691, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.14094955489614244, | |
| "grad_norm": 0.381939560174942, | |
| "learning_rate": 0.00016527444776789915, | |
| "loss": 0.4569, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.14836795252225518, | |
| "grad_norm": 0.37929609417915344, | |
| "learning_rate": 0.00015900000000000002, | |
| "loss": 0.4476, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14836795252225518, | |
| "eval_loss": 0.44979003071784973, | |
| "eval_runtime": 63.1627, | |
| "eval_samples_per_second": 8.993, | |
| "eval_steps_per_second": 2.248, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15578635014836795, | |
| "grad_norm": 0.3976655602455139, | |
| "learning_rate": 0.0001524673415596422, | |
| "loss": 0.4856, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.1632047477744807, | |
| "grad_norm": 0.38872918486595154, | |
| "learning_rate": 0.00014570829890208668, | |
| "loss": 0.4764, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.17062314540059348, | |
| "grad_norm": 0.43565240502357483, | |
| "learning_rate": 0.00013875580140374443, | |
| "loss": 0.4344, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.17804154302670624, | |
| "grad_norm": 0.3594364821910858, | |
| "learning_rate": 0.00013164372093356477, | |
| "loss": 0.4404, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.18545994065281898, | |
| "grad_norm": 0.40855279564857483, | |
| "learning_rate": 0.00012440670683269464, | |
| "loss": 0.4161, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.18545994065281898, | |
| "eval_loss": 0.4417240619659424, | |
| "eval_runtime": 63.0987, | |
| "eval_samples_per_second": 9.002, | |
| "eval_steps_per_second": 2.25, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.19287833827893175, | |
| "grad_norm": 0.4053170084953308, | |
| "learning_rate": 0.00011708001710637128, | |
| "loss": 0.4592, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.20029673590504452, | |
| "grad_norm": 0.40735432505607605, | |
| "learning_rate": 0.00010969934665046512, | |
| "loss": 0.4535, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.20771513353115728, | |
| "grad_norm": 0.3716582953929901, | |
| "learning_rate": 0.00010230065334953492, | |
| "loss": 0.4357, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.21513353115727002, | |
| "grad_norm": 0.3450051546096802, | |
| "learning_rate": 9.491998289362875e-05, | |
| "loss": 0.3879, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.22255192878338279, | |
| "grad_norm": 0.34627842903137207, | |
| "learning_rate": 8.759329316730539e-05, | |
| "loss": 0.4372, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22255192878338279, | |
| "eval_loss": 0.43453946709632874, | |
| "eval_runtime": 63.1414, | |
| "eval_samples_per_second": 8.996, | |
| "eval_steps_per_second": 2.249, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22997032640949555, | |
| "grad_norm": 0.4011070132255554, | |
| "learning_rate": 8.035627906643523e-05, | |
| "loss": 0.444, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.23738872403560832, | |
| "grad_norm": 0.4104556739330292, | |
| "learning_rate": 7.324419859625559e-05, | |
| "loss": 0.4814, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.24480712166172106, | |
| "grad_norm": 0.39093947410583496, | |
| "learning_rate": 6.629170109791332e-05, | |
| "loss": 0.437, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.2522255192878338, | |
| "grad_norm": 0.34994885325431824, | |
| "learning_rate": 5.9532658440357784e-05, | |
| "loss": 0.3983, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.2596439169139466, | |
| "grad_norm": 0.3946869671344757, | |
| "learning_rate": 5.300000000000002e-05, | |
| "loss": 0.4235, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.2596439169139466, | |
| "eval_loss": 0.4264315962791443, | |
| "eval_runtime": 63.1459, | |
| "eval_samples_per_second": 8.995, | |
| "eval_steps_per_second": 2.249, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.26706231454005935, | |
| "grad_norm": 0.40910935401916504, | |
| "learning_rate": 4.672555223210085e-05, | |
| "loss": 0.4254, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.2744807121661721, | |
| "grad_norm": 0.4382137358188629, | |
| "learning_rate": 4.073988361548022e-05, | |
| "loss": 0.4489, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.2818991097922849, | |
| "grad_norm": 0.3514750599861145, | |
| "learning_rate": 3.507215572596106e-05, | |
| "loss": 0.4396, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.2893175074183976, | |
| "grad_norm": 0.3766017258167267, | |
| "learning_rate": 2.9749981164102997e-05, | |
| "loss": 0.3942, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.29673590504451036, | |
| "grad_norm": 0.3943521976470947, | |
| "learning_rate": 2.479928902938834e-05, | |
| "loss": 0.3899, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29673590504451036, | |
| "eval_loss": 0.42192989587783813, | |
| "eval_runtime": 63.1588, | |
| "eval_samples_per_second": 8.993, | |
| "eval_steps_per_second": 2.248, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.30415430267062316, | |
| "grad_norm": 0.39846813678741455, | |
| "learning_rate": 2.024419859625558e-05, | |
| "loss": 0.4341, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.3115727002967359, | |
| "grad_norm": 0.351679265499115, | |
| "learning_rate": 1.610690180741885e-05, | |
| "loss": 0.4537, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.3189910979228487, | |
| "grad_norm": 0.3849548399448395, | |
| "learning_rate": 1.240755515695374e-05, | |
| "loss": 0.3926, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.3264094955489614, | |
| "grad_norm": 0.34369608759880066, | |
| "learning_rate": 9.164181489884296e-06, | |
| "loss": 0.3969, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.33382789317507416, | |
| "grad_norm": 0.3698439300060272, | |
| "learning_rate": 6.392582196693718e-06, | |
| "loss": 0.3801, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.33382789317507416, | |
| "eval_loss": 0.4198478162288666, | |
| "eval_runtime": 63.1389, | |
| "eval_samples_per_second": 8.996, | |
| "eval_steps_per_second": 2.249, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.34124629080118696, | |
| "grad_norm": 0.37923330068588257, | |
| "learning_rate": 4.106260230538197e-06, | |
| "loss": 0.4221, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.3486646884272997, | |
| "grad_norm": 0.3553723990917206, | |
| "learning_rate": 2.316354322216597e-06, | |
| "loss": 0.4364, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.3560830860534125, | |
| "grad_norm": 0.38167324662208557, | |
| "learning_rate": 1.0315847133935416e-06, | |
| "loss": 0.4317, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.36350148367952523, | |
| "grad_norm": 0.36284124851226807, | |
| "learning_rate": 2.582106724586351e-07, | |
| "loss": 0.4302, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.37091988130563797, | |
| "grad_norm": 0.3405497074127197, | |
| "learning_rate": 0.0, | |
| "loss": 0.3725, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37091988130563797, | |
| "eval_loss": 0.4194607138633728, | |
| "eval_runtime": 63.1755, | |
| "eval_samples_per_second": 8.991, | |
| "eval_steps_per_second": 2.248, | |
| "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": 1.85471820890112e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |