| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.25985275010827197, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0034647033347769596, | |
| "grad_norm": 71.73021697998047, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 56.7725, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0034647033347769596, | |
| "eval_loss": 5.7633891105651855, | |
| "eval_runtime": 54.7107, | |
| "eval_samples_per_second": 4.46, | |
| "eval_steps_per_second": 2.23, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006929406669553919, | |
| "grad_norm": 58.02839279174805, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 45.1267, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01039411000433088, | |
| "grad_norm": 62.47406768798828, | |
| "learning_rate": 0.0002, | |
| "loss": 51.5264, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013858813339107838, | |
| "grad_norm": 50.805538177490234, | |
| "learning_rate": 0.0001999048221581858, | |
| "loss": 48.1345, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0173235166738848, | |
| "grad_norm": 53.90876388549805, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 33.368, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02078822000866176, | |
| "grad_norm": 47.89884567260742, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 20.7941, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.024252923343438718, | |
| "grad_norm": 46.44947814941406, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 19.2011, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.027717626678215677, | |
| "grad_norm": 29.739330291748047, | |
| "learning_rate": 0.00019762960071199333, | |
| "loss": 11.6099, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.031182330012992636, | |
| "grad_norm": 37.299049377441406, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 10.8285, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0346470333477696, | |
| "grad_norm": 37.623783111572266, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 8.8167, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.038111736682546554, | |
| "grad_norm": 25.743181228637695, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 7.6365, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04157644001732352, | |
| "grad_norm": 21.669662475585938, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 6.3246, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04504114335210048, | |
| "grad_norm": 21.14327049255371, | |
| "learning_rate": 0.000190630778703665, | |
| "loss": 3.8779, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.048505846686877435, | |
| "grad_norm": 32.12599563598633, | |
| "learning_rate": 0.00018870108331782217, | |
| "loss": 5.3825, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0519705500216544, | |
| "grad_norm": 56.678955078125, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 5.0414, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.055435253356431353, | |
| "grad_norm": 51.66940689086914, | |
| "learning_rate": 0.0001843391445812886, | |
| "loss": 8.305, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.058899956691208316, | |
| "grad_norm": 27.637439727783203, | |
| "learning_rate": 0.0001819152044288992, | |
| "loss": 6.1433, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06236466002598527, | |
| "grad_norm": 28.653242111206055, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 4.0016, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06582936336076224, | |
| "grad_norm": 23.004844665527344, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 4.7562, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0692940666955392, | |
| "grad_norm": 25.145166397094727, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 4.5456, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07275877003031615, | |
| "grad_norm": 16.525177001953125, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 4.3994, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07622347336509311, | |
| "grad_norm": 21.258804321289062, | |
| "learning_rate": 0.00016755902076156604, | |
| "loss": 4.9417, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.07968817669987008, | |
| "grad_norm": 18.620487213134766, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 4.2276, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08315288003464703, | |
| "grad_norm": 71.73140716552734, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 5.1532, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08661758336942399, | |
| "grad_norm": 32.46612548828125, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 3.328, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08661758336942399, | |
| "eval_loss": 0.6132519841194153, | |
| "eval_runtime": 55.2353, | |
| "eval_samples_per_second": 4.417, | |
| "eval_steps_per_second": 2.209, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09008228670420096, | |
| "grad_norm": 30.413021087646484, | |
| "learning_rate": 0.0001537299608346824, | |
| "loss": 4.3897, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.09354699003897791, | |
| "grad_norm": 70.32632446289062, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 6.1758, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09701169337375487, | |
| "grad_norm": 44.74322509765625, | |
| "learning_rate": 0.00014617486132350343, | |
| "loss": 5.2728, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.10047639670853183, | |
| "grad_norm": 44.46076965332031, | |
| "learning_rate": 0.00014226182617406996, | |
| "loss": 7.1594, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.1039411000433088, | |
| "grad_norm": 55.632720947265625, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 3.2506, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10740580337808575, | |
| "grad_norm": 64.20110321044922, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 4.082, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.11087050671286271, | |
| "grad_norm": 82.19364166259766, | |
| "learning_rate": 0.00013007057995042732, | |
| "loss": 5.7286, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.11433521004763968, | |
| "grad_norm": 25.911882400512695, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 3.3962, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.11779991338241663, | |
| "grad_norm": 34.44158935546875, | |
| "learning_rate": 0.00012164396139381029, | |
| "loss": 4.11, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.12126461671719359, | |
| "grad_norm": 29.16976547241211, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 4.2445, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.12472932005197054, | |
| "grad_norm": 18.717464447021484, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 3.1105, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1281940233867475, | |
| "grad_norm": 12.39698314666748, | |
| "learning_rate": 0.00010871557427476583, | |
| "loss": 2.9836, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.13165872672152448, | |
| "grad_norm": 20.110858917236328, | |
| "learning_rate": 0.00010436193873653361, | |
| "loss": 4.8843, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.13512343005630142, | |
| "grad_norm": 22.39121437072754, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7531, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1385881333910784, | |
| "grad_norm": 12.983887672424316, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 2.8063, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14205283672585534, | |
| "grad_norm": 12.021308898925781, | |
| "learning_rate": 9.128442572523417e-05, | |
| "loss": 1.8465, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.1455175400606323, | |
| "grad_norm": 32.44965744018555, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 4.7937, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.14898224339540928, | |
| "grad_norm": 16.291828155517578, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 4.0307, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.15244694673018622, | |
| "grad_norm": 12.486902236938477, | |
| "learning_rate": 7.835603860618972e-05, | |
| "loss": 3.1398, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1559116500649632, | |
| "grad_norm": 20.737529754638672, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 2.7772, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15937635339974016, | |
| "grad_norm": 17.130455017089844, | |
| "learning_rate": 6.992942004957271e-05, | |
| "loss": 2.7659, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1628410567345171, | |
| "grad_norm": 33.27773666381836, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 4.4388, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.16630576006929407, | |
| "grad_norm": 16.671640396118164, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 5.4897, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.16977046340407104, | |
| "grad_norm": 32.55076599121094, | |
| "learning_rate": 5.773817382593008e-05, | |
| "loss": 5.7507, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.17323516673884798, | |
| "grad_norm": 24.630674362182617, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 4.7681, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17323516673884798, | |
| "eval_loss": 0.4547581374645233, | |
| "eval_runtime": 55.2618, | |
| "eval_samples_per_second": 4.415, | |
| "eval_steps_per_second": 2.208, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17669987007362495, | |
| "grad_norm": 35.518733978271484, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 3.7489, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.18016457340840192, | |
| "grad_norm": 13.69632625579834, | |
| "learning_rate": 4.6270039165317605e-05, | |
| "loss": 3.3506, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.18362927674317886, | |
| "grad_norm": 8.78832721710205, | |
| "learning_rate": 4.264235636489542e-05, | |
| "loss": 1.1164, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.18709398007795583, | |
| "grad_norm": 19.308809280395508, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 2.4321, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.19055868341273277, | |
| "grad_norm": 37.123531341552734, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 2.8694, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.19402338674750974, | |
| "grad_norm": 17.861862182617188, | |
| "learning_rate": 3.244097923843398e-05, | |
| "loss": 2.7996, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1974880900822867, | |
| "grad_norm": 25.85512351989746, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 3.5098, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.20095279341706365, | |
| "grad_norm": 15.257984161376953, | |
| "learning_rate": 2.6272266318987603e-05, | |
| "loss": 4.2756, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.20441749675184062, | |
| "grad_norm": 9.036160469055176, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.1725, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.2078822000866176, | |
| "grad_norm": 8.915935516357422, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 2.2972, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21134690342139453, | |
| "grad_norm": 21.377731323242188, | |
| "learning_rate": 1.808479557110081e-05, | |
| "loss": 3.7501, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.2148116067561715, | |
| "grad_norm": 25.860002517700195, | |
| "learning_rate": 1.566085541871145e-05, | |
| "loss": 4.4689, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.21827631009094847, | |
| "grad_norm": 18.97432518005371, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 1.9634, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.22174101342572541, | |
| "grad_norm": 19.91278839111328, | |
| "learning_rate": 1.129891668217783e-05, | |
| "loss": 2.4683, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.22520571676050238, | |
| "grad_norm": 23.109867095947266, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 5.7752, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.22867042009527935, | |
| "grad_norm": 15.709304809570312, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 3.541, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.2321351234300563, | |
| "grad_norm": 20.902395248413086, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 3.8023, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.23559982676483326, | |
| "grad_norm": 23.645151138305664, | |
| "learning_rate": 4.628304925177318e-06, | |
| "loss": 3.0177, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.23906453009961023, | |
| "grad_norm": 17.126794815063477, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 2.0654, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.24252923343438718, | |
| "grad_norm": 20.345962524414062, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 3.7378, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24599393676916415, | |
| "grad_norm": 21.93646812438965, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 2.4299, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.2494586401039411, | |
| "grad_norm": 20.879924774169922, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 2.5581, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2529233434387181, | |
| "grad_norm": 31.036603927612305, | |
| "learning_rate": 3.805301908254455e-07, | |
| "loss": 4.7284, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.256388046773495, | |
| "grad_norm": 24.942869186401367, | |
| "learning_rate": 9.517784181422019e-08, | |
| "loss": 2.1838, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.25985275010827197, | |
| "grad_norm": 11.447266578674316, | |
| "learning_rate": 0.0, | |
| "loss": 2.1105, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.25985275010827197, | |
| "eval_loss": 0.38869383931159973, | |
| "eval_runtime": 55.2151, | |
| "eval_samples_per_second": 4.419, | |
| "eval_steps_per_second": 2.21, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9043369091072e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |