| { | |
| "best_metric": 0.7143663763999939, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-4/checkpoint-187", | |
| "epoch": 40.90909090909091, | |
| "eval_steps": 500, | |
| "global_step": 225, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.7671908140182495, | |
| "learning_rate": 1.3333333333333336e-07, | |
| "loss": 1.1716, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.6510481834411621, | |
| "learning_rate": 2.666666666666667e-07, | |
| "loss": 1.0331, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.702520489692688, | |
| "learning_rate": 5.333333333333335e-07, | |
| "loss": 1.1544, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.1236884593963623, | |
| "eval_runtime": 24.3567, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.7136379480361938, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.0735, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.7056083083152771, | |
| "learning_rate": 1.066666666666667e-06, | |
| "loss": 1.1336, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.7233437895774841, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.132, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1193286180496216, | |
| "eval_runtime": 24.3549, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.6547144055366516, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 1.0473, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.7151400446891785, | |
| "learning_rate": 1.8666666666666669e-06, | |
| "loss": 1.1478, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.7789710760116577, | |
| "learning_rate": 2.133333333333334e-06, | |
| "loss": 1.0689, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 1.1136311292648315, | |
| "eval_runtime": 24.3553, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.6943601369857788, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.1271, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.691636860370636, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.0717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6316518187522888, | |
| "learning_rate": 2.9333333333333338e-06, | |
| "loss": 1.0956, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1010679006576538, | |
| "eval_runtime": 24.358, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.7023854851722717, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.1054, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.7613805532455444, | |
| "learning_rate": 3.4666666666666672e-06, | |
| "loss": 1.1157, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 1.087128758430481, | |
| "eval_runtime": 24.3561, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.6336562633514404, | |
| "learning_rate": 3.7333333333333337e-06, | |
| "loss": 0.9922, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.7272669076919556, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0947, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.7287123203277588, | |
| "learning_rate": 4.266666666666668e-06, | |
| "loss": 1.0778, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.063869833946228, | |
| "eval_runtime": 24.3649, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.6734246015548706, | |
| "learning_rate": 4.533333333333334e-06, | |
| "loss": 1.038, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.628820538520813, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.03, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.6829433441162109, | |
| "learning_rate": 5.0666666666666676e-06, | |
| "loss": 1.0458, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 1.039298176765442, | |
| "eval_runtime": 24.363, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.684455931186676, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.0333, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.6354673504829407, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.9913, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5571680665016174, | |
| "learning_rate": 5.8666666666666675e-06, | |
| "loss": 0.9854, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.0027111768722534, | |
| "eval_runtime": 24.3512, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.5560793876647949, | |
| "learning_rate": 6.133333333333334e-06, | |
| "loss": 0.9371, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.6101187467575073, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.9996, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.9696131348609924, | |
| "eval_runtime": 24.3608, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.5255017280578613, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.9569, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.5077732801437378, | |
| "learning_rate": 6.9333333333333344e-06, | |
| "loss": 0.9602, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.4885198771953583, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.8991, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.9317263960838318, | |
| "eval_runtime": 24.3522, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.42795872688293457, | |
| "learning_rate": 7.4666666666666675e-06, | |
| "loss": 0.9282, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.3717321455478668, | |
| "learning_rate": 7.733333333333334e-06, | |
| "loss": 0.9173, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.37956222891807556, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8897, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.9051777720451355, | |
| "eval_runtime": 24.3554, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.32997962832450867, | |
| "learning_rate": 8.266666666666667e-06, | |
| "loss": 0.8511, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.3067737817764282, | |
| "learning_rate": 8.533333333333335e-06, | |
| "loss": 0.8714, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.3460928797721863, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.8711, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.878821074962616, | |
| "eval_runtime": 24.3541, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.29182004928588867, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.8274, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.31693214178085327, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.8809, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 0.8587873578071594, | |
| "eval_runtime": 24.3566, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.29670482873916626, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.8263, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.2604750096797943, | |
| "learning_rate": 9.866666666666668e-06, | |
| "loss": 0.8237, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.2680836021900177, | |
| "learning_rate": 9.999945845889795e-06, | |
| "loss": 0.7972, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.8367733955383301, | |
| "eval_runtime": 24.3654, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.2775014638900757, | |
| "learning_rate": 9.999512620046523e-06, | |
| "loss": 0.8129, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.24960099160671234, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.8163, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.2805323004722595, | |
| "learning_rate": 9.99734667851357e-06, | |
| "loss": 0.8156, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 0.8207753300666809, | |
| "eval_runtime": 24.3571, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 15.272727272727273, | |
| "grad_norm": 0.23699168860912323, | |
| "learning_rate": 9.995614150494293e-06, | |
| "loss": 0.7658, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 15.636363636363637, | |
| "grad_norm": 0.24089692533016205, | |
| "learning_rate": 9.993448771956285e-06, | |
| "loss": 0.8111, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.24843314290046692, | |
| "learning_rate": 9.99085073052117e-06, | |
| "loss": 0.7815, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8057171702384949, | |
| "eval_runtime": 24.3555, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.363636363636363, | |
| "grad_norm": 0.2455184906721115, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.7832, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.727272727272727, | |
| "grad_norm": 0.23427604138851166, | |
| "learning_rate": 9.984357596869369e-06, | |
| "loss": 0.7492, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.90909090909091, | |
| "eval_loss": 0.7956312298774719, | |
| "eval_runtime": 24.3597, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 17.09090909090909, | |
| "grad_norm": 0.2534084618091583, | |
| "learning_rate": 9.980463067257437e-06, | |
| "loss": 0.774, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 17.454545454545453, | |
| "grad_norm": 0.2455044388771057, | |
| "learning_rate": 9.976136999909156e-06, | |
| "loss": 0.7431, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 17.818181818181817, | |
| "grad_norm": 0.22460192441940308, | |
| "learning_rate": 9.971379769661422e-06, | |
| "loss": 0.7587, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.7855010628700256, | |
| "eval_runtime": 24.3575, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "grad_norm": 0.2319430559873581, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.7886, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 18.545454545454547, | |
| "grad_norm": 0.22566601634025574, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 0.7468, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "grad_norm": 0.2197207510471344, | |
| "learning_rate": 9.95452541005172e-06, | |
| "loss": 0.7483, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "eval_loss": 0.7779585719108582, | |
| "eval_runtime": 24.351, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 19.272727272727273, | |
| "grad_norm": 0.25423386693000793, | |
| "learning_rate": 9.948048023191728e-06, | |
| "loss": 0.7066, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 19.636363636363637, | |
| "grad_norm": 0.27388209104537964, | |
| "learning_rate": 9.941141907232766e-06, | |
| "loss": 0.7644, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.21174727380275726, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.7296, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7694642543792725, | |
| "eval_runtime": 24.3585, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.363636363636363, | |
| "grad_norm": 0.21564796566963196, | |
| "learning_rate": 9.926045918666045e-06, | |
| "loss": 0.7005, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 20.727272727272727, | |
| "grad_norm": 0.21516019105911255, | |
| "learning_rate": 9.91785735406693e-06, | |
| "loss": 0.7441, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 20.90909090909091, | |
| "eval_loss": 0.7629389762878418, | |
| "eval_runtime": 24.3562, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 21.09090909090909, | |
| "grad_norm": 0.25928786396980286, | |
| "learning_rate": 9.909242676272797e-06, | |
| "loss": 0.7448, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 21.454545454545453, | |
| "grad_norm": 0.22024983167648315, | |
| "learning_rate": 9.90020263171194e-06, | |
| "loss": 0.7277, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 21.818181818181817, | |
| "grad_norm": 0.22692517936229706, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.7176, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.7560660243034363, | |
| "eval_runtime": 24.3539, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 22.181818181818183, | |
| "grad_norm": 0.2096530646085739, | |
| "learning_rate": 9.880849612217238e-06, | |
| "loss": 0.6737, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 22.545454545454547, | |
| "grad_norm": 0.23917944729328156, | |
| "learning_rate": 9.870538314147194e-06, | |
| "loss": 0.7373, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 22.90909090909091, | |
| "grad_norm": 0.2294284701347351, | |
| "learning_rate": 9.859805002892733e-06, | |
| "loss": 0.7033, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 22.90909090909091, | |
| "eval_loss": 0.7507641911506653, | |
| "eval_runtime": 24.3597, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 23.272727272727273, | |
| "grad_norm": 0.22704288363456726, | |
| "learning_rate": 9.84865060845349e-06, | |
| "loss": 0.6932, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 23.636363636363637, | |
| "grad_norm": 0.2241230458021164, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.6907, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.2692793011665344, | |
| "learning_rate": 9.825082472361558e-06, | |
| "loss": 0.6906, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.7443225383758545, | |
| "eval_runtime": 24.3494, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 24.363636363636363, | |
| "grad_norm": 0.2639993727207184, | |
| "learning_rate": 9.812670772796113e-06, | |
| "loss": 0.7003, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 24.727272727272727, | |
| "grad_norm": 0.2364564836025238, | |
| "learning_rate": 9.799842074043438e-06, | |
| "loss": 0.6954, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 24.90909090909091, | |
| "eval_loss": 0.7396196722984314, | |
| "eval_runtime": 24.3588, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 25.09090909090909, | |
| "grad_norm": 0.25801482796669006, | |
| "learning_rate": 9.786597487660336e-06, | |
| "loss": 0.6571, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 25.454545454545453, | |
| "grad_norm": 0.23688581585884094, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.6904, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 25.818181818181817, | |
| "grad_norm": 0.2453926056623459, | |
| "learning_rate": 9.75886527830587e-06, | |
| "loss": 0.6578, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.734368622303009, | |
| "eval_runtime": 24.3632, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 26.181818181818183, | |
| "grad_norm": 0.2570464611053467, | |
| "learning_rate": 9.744380058222483e-06, | |
| "loss": 0.6708, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 26.545454545454547, | |
| "grad_norm": 0.2480405867099762, | |
| "learning_rate": 9.729483756076436e-06, | |
| "loss": 0.6975, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 26.90909090909091, | |
| "grad_norm": 0.2474363148212433, | |
| "learning_rate": 9.714177662574316e-06, | |
| "loss": 0.6495, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 26.90909090909091, | |
| "eval_loss": 0.7309517860412598, | |
| "eval_runtime": 24.351, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 27.272727272727273, | |
| "grad_norm": 0.2587747871875763, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.6405, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 27.636363636363637, | |
| "grad_norm": 0.2785438597202301, | |
| "learning_rate": 9.682341441747446e-06, | |
| "loss": 0.6908, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.25493577122688293, | |
| "learning_rate": 9.665814072907293e-06, | |
| "loss": 0.6391, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.7269113063812256, | |
| "eval_runtime": 24.36, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 28.363636363636363, | |
| "grad_norm": 0.25702783465385437, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 0.6446, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 28.727272727272727, | |
| "grad_norm": 0.2598085105419159, | |
| "learning_rate": 9.63154797841033e-06, | |
| "loss": 0.6442, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 28.90909090909091, | |
| "eval_loss": 0.723728358745575, | |
| "eval_runtime": 24.3516, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 29.09090909090909, | |
| "grad_norm": 0.27804017066955566, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.6514, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 29.454545454545453, | |
| "grad_norm": 0.2686798572540283, | |
| "learning_rate": 9.595676696276173e-06, | |
| "loss": 0.6613, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 29.818181818181817, | |
| "grad_norm": 0.275974303483963, | |
| "learning_rate": 9.577142973279896e-06, | |
| "loss": 0.6268, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.7199351787567139, | |
| "eval_runtime": 24.3608, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 30.181818181818183, | |
| "grad_norm": 0.2688060700893402, | |
| "learning_rate": 9.55821265866333e-06, | |
| "loss": 0.6002, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 30.545454545454547, | |
| "grad_norm": 0.27310389280319214, | |
| "learning_rate": 9.538887392664544e-06, | |
| "loss": 0.6103, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 30.90909090909091, | |
| "grad_norm": 0.294357031583786, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.6536, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 30.90909090909091, | |
| "eval_loss": 0.71826171875, | |
| "eval_runtime": 24.3503, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 31.272727272727273, | |
| "grad_norm": 0.2845761775970459, | |
| "learning_rate": 9.499058738432492e-06, | |
| "loss": 0.6384, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 31.636363636363637, | |
| "grad_norm": 0.30121341347694397, | |
| "learning_rate": 9.478558801197065e-06, | |
| "loss": 0.6191, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.2854570746421814, | |
| "learning_rate": 9.457670814276083e-06, | |
| "loss": 0.6092, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.7163482308387756, | |
| "eval_runtime": 24.3583, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 32.36363636363637, | |
| "grad_norm": 0.3030169904232025, | |
| "learning_rate": 9.436396587532297e-06, | |
| "loss": 0.5929, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 32.72727272727273, | |
| "grad_norm": 0.2960076928138733, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.621, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 32.90909090909091, | |
| "eval_loss": 0.7148606181144714, | |
| "eval_runtime": 24.3579, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 33.09090909090909, | |
| "grad_norm": 0.3322046995162964, | |
| "learning_rate": 9.392696821198488e-06, | |
| "loss": 0.6048, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 33.45454545454545, | |
| "grad_norm": 0.335716187953949, | |
| "learning_rate": 9.370275068023097e-06, | |
| "loss": 0.6233, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 33.81818181818182, | |
| "grad_norm": 0.3075459599494934, | |
| "learning_rate": 9.347474647526095e-06, | |
| "loss": 0.5823, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.7143663763999939, | |
| "eval_runtime": 24.3571, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 34.18181818181818, | |
| "grad_norm": 0.3422604203224182, | |
| "learning_rate": 9.324297535275156e-06, | |
| "loss": 0.5754, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 34.54545454545455, | |
| "grad_norm": 0.33866289258003235, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.5951, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 34.90909090909091, | |
| "grad_norm": 0.37743914127349854, | |
| "learning_rate": 9.276821300802535e-06, | |
| "loss": 0.5651, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 34.90909090909091, | |
| "eval_loss": 0.7156122326850891, | |
| "eval_runtime": 24.3497, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 35.27272727272727, | |
| "grad_norm": 0.3403712511062622, | |
| "learning_rate": 9.25252629221175e-06, | |
| "loss": 0.5877, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 35.63636363636363, | |
| "grad_norm": 0.32227441668510437, | |
| "learning_rate": 9.227862818772392e-06, | |
| "loss": 0.5569, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.3595390021800995, | |
| "learning_rate": 9.202833017478421e-06, | |
| "loss": 0.5951, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.7163762450218201, | |
| "eval_runtime": 24.3582, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 36.36363636363637, | |
| "grad_norm": 0.3400909900665283, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.552, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 36.72727272727273, | |
| "grad_norm": 0.3983656167984009, | |
| "learning_rate": 9.151683137818989e-06, | |
| "loss": 0.5637, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 36.90909090909091, | |
| "eval_loss": 0.7194623947143555, | |
| "eval_runtime": 24.3553, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 37.09090909090909, | |
| "grad_norm": 0.4176684319972992, | |
| "learning_rate": 9.125567491391476e-06, | |
| "loss": 0.5684, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 37.45454545454545, | |
| "grad_norm": 0.3539072275161743, | |
| "learning_rate": 9.099094380601244e-06, | |
| "loss": 0.5392, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 37.81818181818182, | |
| "grad_norm": 0.40454044938087463, | |
| "learning_rate": 9.072266099240286e-06, | |
| "loss": 0.5669, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.7218795418739319, | |
| "eval_runtime": 24.3543, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 38.18181818181818, | |
| "grad_norm": 0.3737054765224457, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.5348, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 38.54545454545455, | |
| "grad_norm": 0.37676355242729187, | |
| "learning_rate": 9.017553353643479e-06, | |
| "loss": 0.5322, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 38.90909090909091, | |
| "grad_norm": 0.41781264543533325, | |
| "learning_rate": 8.989673630054044e-06, | |
| "loss": 0.5613, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 38.90909090909091, | |
| "eval_loss": 0.7278082966804504, | |
| "eval_runtime": 24.3537, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 39.27272727272727, | |
| "grad_norm": 0.41486313939094543, | |
| "learning_rate": 8.961448216775955e-06, | |
| "loss": 0.5221, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 39.63636363636363, | |
| "grad_norm": 0.40721336007118225, | |
| "learning_rate": 8.932879559431392e-06, | |
| "loss": 0.5181, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.46421483159065247, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 0.5156, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.7309436798095703, | |
| "eval_runtime": 24.3653, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 40.36363636363637, | |
| "grad_norm": 0.5447797179222107, | |
| "learning_rate": 8.874722443520898e-06, | |
| "loss": 0.5196, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 40.72727272727273, | |
| "grad_norm": 0.44560515880584717, | |
| "learning_rate": 8.845139024042664e-06, | |
| "loss": 0.5044, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 40.90909090909091, | |
| "eval_loss": 0.739515483379364, | |
| "eval_runtime": 24.3585, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 40.90909090909091, | |
| "step": 225, | |
| "total_flos": 3.429455470966866e+17, | |
| "train_loss": 0.7736989471647474, | |
| "train_runtime": 12307.7975, | |
| "train_samples_per_second": 1.072, | |
| "train_steps_per_second": 0.061 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "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": 3.429455470966866e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |