| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.5599553820412715, |
| "eval_steps": 100, |
| "global_step": 700, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.022308979364194088, |
| "grad_norm": 4.709710597991943, |
| "learning_rate": 0.00019999446787180336, |
| "loss": 3.676974868774414, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.044617958728388175, |
| "grad_norm": 1.4567488431930542, |
| "learning_rate": 0.00019989613595281384, |
| "loss": 0.9999975204467774, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.06692693809258227, |
| "grad_norm": 1.1962672472000122, |
| "learning_rate": 0.00019967500698688952, |
| "loss": 0.6824670314788819, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.08923591745677635, |
| "grad_norm": 0.8628137707710266, |
| "learning_rate": 0.0001993313527961959, |
| "loss": 0.5993218421936035, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.11154489682097044, |
| "grad_norm": 1.0308934450149536, |
| "learning_rate": 0.00019886559581668999, |
| "loss": 0.542162561416626, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.13385387618516453, |
| "grad_norm": 0.9261613488197327, |
| "learning_rate": 0.00019827830857884173, |
| "loss": 0.44734554290771483, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.1561628555493586, |
| "grad_norm": 0.9318423867225647, |
| "learning_rate": 0.00019757021300385286, |
| "loss": 0.44412760734558104, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.1784718349135527, |
| "grad_norm": 0.5784896612167358, |
| "learning_rate": 0.00019674217951623707, |
| "loss": 0.40378632545471194, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.2007808142777468, |
| "grad_norm": 0.7836484909057617, |
| "learning_rate": 0.00019579522597385315, |
| "loss": 0.3856403350830078, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.22308979364194087, |
| "grad_norm": 0.6503493189811707, |
| "learning_rate": 0.00019473051641670606, |
| "loss": 0.3800614356994629, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.22308979364194087, |
| "eval_loss": 8.158984184265137, |
| "eval_runtime": 248.821, |
| "eval_samples_per_second": 1.813, |
| "eval_steps_per_second": 1.813, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.24539877300613497, |
| "grad_norm": 0.582433819770813, |
| "learning_rate": 0.00019354935963605393, |
| "loss": 0.34090685844421387, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.26770775237032907, |
| "grad_norm": 0.6219270825386047, |
| "learning_rate": 0.00019225320756558023, |
| "loss": 0.3560009956359863, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.29001673173452314, |
| "grad_norm": 0.5081164836883545, |
| "learning_rate": 0.0001908436534966081, |
| "loss": 0.32209837436676025, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.3123257110987172, |
| "grad_norm": 1.0672804117202759, |
| "learning_rate": 0.00018932243011955154, |
| "loss": 0.4280549049377441, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.33463469046291133, |
| "grad_norm": 0.5716775059700012, |
| "learning_rate": 0.00018769140739401062, |
| "loss": 0.29103860855102537, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.3569436698271054, |
| "grad_norm": 0.6793851256370544, |
| "learning_rate": 0.00018595259025012877, |
| "loss": 0.3713259696960449, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.3792526491912995, |
| "grad_norm": 0.6091159582138062, |
| "learning_rate": 0.0001841081161240379, |
| "loss": 0.3802988529205322, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.4015616285554936, |
| "grad_norm": 0.7563126087188721, |
| "learning_rate": 0.0001821602523304211, |
| "loss": 0.3028958082199097, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.42387060791968767, |
| "grad_norm": 0.5827437043190002, |
| "learning_rate": 0.00018011139327542237, |
| "loss": 0.3516202211380005, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.44617958728388174, |
| "grad_norm": 0.6265884041786194, |
| "learning_rate": 0.0001779640575133296, |
| "loss": 0.3505818843841553, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.44617958728388174, |
| "eval_loss": 6.77685546875, |
| "eval_runtime": 246.6844, |
| "eval_samples_per_second": 1.828, |
| "eval_steps_per_second": 1.828, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.46848856664807587, |
| "grad_norm": 0.4515160024166107, |
| "learning_rate": 0.00017572088465064848, |
| "loss": 0.30297350883483887, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.49079754601226994, |
| "grad_norm": 0.47408005595207214, |
| "learning_rate": 0.0001733846321013738, |
| "loss": 0.30213174819946287, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.5131065253764641, |
| "grad_norm": 0.43776413798332214, |
| "learning_rate": 0.00017095817169744595, |
| "loss": 0.2981758117675781, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.5354155047406581, |
| "grad_norm": 0.5996765494346619, |
| "learning_rate": 0.00016844448615855933, |
| "loss": 0.2770395278930664, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.5577244841048522, |
| "grad_norm": 1.4966278076171875, |
| "learning_rate": 0.0001658466654256627, |
| "loss": 0.3032949924468994, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.5800334634690463, |
| "grad_norm": 0.5301216840744019, |
| "learning_rate": 0.00016316790286265763, |
| "loss": 0.3215930700302124, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.6023424428332403, |
| "grad_norm": 0.5374172925949097, |
| "learning_rate": 0.00016041149133096512, |
| "loss": 0.29086947441101074, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.6246514221974344, |
| "grad_norm": 0.5407304167747498, |
| "learning_rate": 0.00015758081914178456, |
| "loss": 0.26963791847229, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.6469604015616286, |
| "grad_norm": 0.40757226943969727, |
| "learning_rate": 0.00015467936589102176, |
| "loss": 0.3320644378662109, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.6692693809258227, |
| "grad_norm": 0.36359116435050964, |
| "learning_rate": 0.00015171069818200548, |
| "loss": 0.28259494304656985, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6692693809258227, |
| "eval_loss": 6.212057590484619, |
| "eval_runtime": 246.7095, |
| "eval_samples_per_second": 1.828, |
| "eval_steps_per_second": 1.828, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6915783602900167, |
| "grad_norm": 0.508072018623352, |
| "learning_rate": 0.00014867846524125, |
| "loss": 0.29235794544219973, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.7138873396542108, |
| "grad_norm": 0.7422248125076294, |
| "learning_rate": 0.0001455863944326538, |
| "loss": 0.3195211887359619, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.7361963190184049, |
| "grad_norm": 0.6975764632225037, |
| "learning_rate": 0.00014243828667564768, |
| "loss": 0.30194265842437745, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.758505298382599, |
| "grad_norm": 0.607437014579773, |
| "learning_rate": 0.00013923801177292527, |
| "loss": 0.2903664350509644, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.7808142777467931, |
| "grad_norm": 0.5284196734428406, |
| "learning_rate": 0.00013598950365349883, |
| "loss": 0.2615989923477173, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.8031232571109872, |
| "grad_norm": 0.44667983055114746, |
| "learning_rate": 0.00013269675553692785, |
| "loss": 0.32799758911132815, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.8254322364751813, |
| "grad_norm": 0.440305233001709, |
| "learning_rate": 0.00012936381502466523, |
| "loss": 0.28976309299468994, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.8477412158393753, |
| "grad_norm": 0.5174661874771118, |
| "learning_rate": 0.00012599477912455425, |
| "loss": 0.2578805685043335, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.8700501952035694, |
| "grad_norm": 0.33937904238700867, |
| "learning_rate": 0.0001225937892145932, |
| "loss": 0.27044386863708497, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.8923591745677635, |
| "grad_norm": 0.5235030651092529, |
| "learning_rate": 0.000119165025952158, |
| "loss": 0.2892023801803589, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.8923591745677635, |
| "eval_loss": 5.747542381286621, |
| "eval_runtime": 249.1726, |
| "eval_samples_per_second": 1.81, |
| "eval_steps_per_second": 1.81, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.9146681539319577, |
| "grad_norm": 0.4365766942501068, |
| "learning_rate": 0.00011571270413494082, |
| "loss": 0.28197059631347654, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.9369771332961517, |
| "grad_norm": 0.42031630873680115, |
| "learning_rate": 0.00011224106751992163, |
| "loss": 0.2776280641555786, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.9592861126603458, |
| "grad_norm": 0.41079336404800415, |
| "learning_rate": 0.0001087543836067418, |
| "loss": 0.23740179538726808, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.9815950920245399, |
| "grad_norm": 0.6885762810707092, |
| "learning_rate": 0.00010525693839189214, |
| "loss": 0.2687765836715698, |
| "step": 440 |
| }, |
| { |
| "epoch": 1.0022308979364194, |
| "grad_norm": 0.4392392337322235, |
| "learning_rate": 0.00010175303110016343, |
| "loss": 0.3171211004257202, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.0245398773006136, |
| "grad_norm": 0.3510143756866455, |
| "learning_rate": 9.824696889983662e-05, |
| "loss": 0.23552191257476807, |
| "step": 460 |
| }, |
| { |
| "epoch": 1.0468488566648075, |
| "grad_norm": 0.3384598195552826, |
| "learning_rate": 9.474306160810791e-05, |
| "loss": 0.23134019374847412, |
| "step": 470 |
| }, |
| { |
| "epoch": 1.0691578360290017, |
| "grad_norm": 0.40072542428970337, |
| "learning_rate": 9.124561639325822e-05, |
| "loss": 0.23302733898162842, |
| "step": 480 |
| }, |
| { |
| "epoch": 1.0914668153931957, |
| "grad_norm": 0.45970481634140015, |
| "learning_rate": 8.775893248007839e-05, |
| "loss": 0.2210915803909302, |
| "step": 490 |
| }, |
| { |
| "epoch": 1.1137757947573899, |
| "grad_norm": 0.4819018840789795, |
| "learning_rate": 8.42872958650592e-05, |
| "loss": 0.22845861911773682, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.1137757947573899, |
| "eval_loss": 5.7599334716796875, |
| "eval_runtime": 243.806, |
| "eval_samples_per_second": 1.85, |
| "eval_steps_per_second": 1.85, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.136084774121584, |
| "grad_norm": 0.3909103572368622, |
| "learning_rate": 8.083497404784201e-05, |
| "loss": 0.253029990196228, |
| "step": 510 |
| }, |
| { |
| "epoch": 1.158393753485778, |
| "grad_norm": 0.49080875515937805, |
| "learning_rate": 7.740621078540684e-05, |
| "loss": 0.22022719383239747, |
| "step": 520 |
| }, |
| { |
| "epoch": 1.1807027328499722, |
| "grad_norm": 0.3979121446609497, |
| "learning_rate": 7.400522087544582e-05, |
| "loss": 0.24559009075164795, |
| "step": 530 |
| }, |
| { |
| "epoch": 1.2030117122141661, |
| "grad_norm": 0.6220426559448242, |
| "learning_rate": 7.063618497533479e-05, |
| "loss": 0.23752431869506835, |
| "step": 540 |
| }, |
| { |
| "epoch": 1.2253206915783603, |
| "grad_norm": 0.3583706319332123, |
| "learning_rate": 6.730324446307217e-05, |
| "loss": 0.23795950412750244, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.2476296709425543, |
| "grad_norm": 0.42460814118385315, |
| "learning_rate": 6.401049634650118e-05, |
| "loss": 0.22870173454284667, |
| "step": 560 |
| }, |
| { |
| "epoch": 1.2699386503067485, |
| "grad_norm": 0.4157378077507019, |
| "learning_rate": 6.076198822707475e-05, |
| "loss": 0.2094573736190796, |
| "step": 570 |
| }, |
| { |
| "epoch": 1.2922476296709426, |
| "grad_norm": 0.3690703511238098, |
| "learning_rate": 5.7561713324352365e-05, |
| "loss": 0.21153016090393068, |
| "step": 580 |
| }, |
| { |
| "epoch": 1.3145566090351366, |
| "grad_norm": 0.3393983542919159, |
| "learning_rate": 5.44136055673462e-05, |
| "loss": 0.23398885726928711, |
| "step": 590 |
| }, |
| { |
| "epoch": 1.3368655883993308, |
| "grad_norm": 0.4122086465358734, |
| "learning_rate": 5.1321534758750037e-05, |
| "loss": 0.2347275733947754, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.3368655883993308, |
| "eval_loss": 5.804348468780518, |
| "eval_runtime": 244.347, |
| "eval_samples_per_second": 1.846, |
| "eval_steps_per_second": 1.846, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.3591745677635247, |
| "grad_norm": 0.4437583386898041, |
| "learning_rate": 4.828930181799457e-05, |
| "loss": 0.2153022050857544, |
| "step": 610 |
| }, |
| { |
| "epoch": 1.381483547127719, |
| "grad_norm": 0.4642995595932007, |
| "learning_rate": 4.532063410897829e-05, |
| "loss": 0.22439846992492676, |
| "step": 620 |
| }, |
| { |
| "epoch": 1.4037925264919129, |
| "grad_norm": 0.7296940088272095, |
| "learning_rate": 4.241918085821547e-05, |
| "loss": 0.2058631181716919, |
| "step": 630 |
| }, |
| { |
| "epoch": 1.426101505856107, |
| "grad_norm": 0.5185000896453857, |
| "learning_rate": 3.9588508669034886e-05, |
| "loss": 0.21068837642669677, |
| "step": 640 |
| }, |
| { |
| "epoch": 1.4484104852203012, |
| "grad_norm": 0.41563302278518677, |
| "learning_rate": 3.683209713734237e-05, |
| "loss": 0.22028052806854248, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.4707194645844952, |
| "grad_norm": 0.2550966739654541, |
| "learning_rate": 3.415333457433733e-05, |
| "loss": 0.2142402172088623, |
| "step": 660 |
| }, |
| { |
| "epoch": 1.4930284439486894, |
| "grad_norm": 0.5091217160224915, |
| "learning_rate": 3.155551384144069e-05, |
| "loss": 0.19590476751327515, |
| "step": 670 |
| }, |
| { |
| "epoch": 1.5153374233128836, |
| "grad_norm": 0.38000059127807617, |
| "learning_rate": 2.9041828302554074e-05, |
| "loss": 0.2405090570449829, |
| "step": 680 |
| }, |
| { |
| "epoch": 1.5376464026770775, |
| "grad_norm": 0.39434534311294556, |
| "learning_rate": 2.661536789862622e-05, |
| "loss": 0.25410382747650145, |
| "step": 690 |
| }, |
| { |
| "epoch": 1.5599553820412715, |
| "grad_norm": 0.449735552072525, |
| "learning_rate": 2.4279115349351543e-05, |
| "loss": 0.21624345779418946, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.5599553820412715, |
| "eval_loss": 5.754471302032471, |
| "eval_runtime": 244.0702, |
| "eval_samples_per_second": 1.848, |
| "eval_steps_per_second": 1.848, |
| "step": 700 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 898, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.132487571153363e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
