| { | |
| "best_metric": 0.8536396622657776, | |
| "best_model_checkpoint": "./lora-alpaca/checkpoint-1200", | |
| "epoch": 2.3139220979560355, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.6999999999999996e-05, | |
| "loss": 1.554, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 1.4793, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.2861, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00011099999999999999, | |
| "loss": 1.1305, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00014099999999999998, | |
| "loss": 1.0653, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00017099999999999998, | |
| "loss": 1.0233, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000201, | |
| "loss": 0.9835, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00023099999999999998, | |
| "loss": 0.9385, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.000261, | |
| "loss": 0.9179, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00029099999999999997, | |
| "loss": 0.9107, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0002985557083906465, | |
| "loss": 0.914, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0002964924346629986, | |
| "loss": 0.8971, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00029442916093535074, | |
| "loss": 0.9021, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00029236588720770286, | |
| "loss": 0.8939, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000290302613480055, | |
| "loss": 0.8917, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0002882393397524071, | |
| "loss": 0.8834, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00028617606602475925, | |
| "loss": 0.8927, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0002841127922971114, | |
| "loss": 0.8979, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0002820495185694635, | |
| "loss": 0.8812, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00027998624484181563, | |
| "loss": 0.8852, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.8834338188171387, | |
| "eval_runtime": 172.9592, | |
| "eval_samples_per_second": 11.563, | |
| "eval_steps_per_second": 0.289, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002779229711141678, | |
| "loss": 0.8837, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00027585969738651994, | |
| "loss": 0.8837, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00027379642365887207, | |
| "loss": 0.8703, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002717331499312242, | |
| "loss": 0.885, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002696698762035763, | |
| "loss": 0.877, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00026760660247592845, | |
| "loss": 0.8698, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002655433287482806, | |
| "loss": 0.8698, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002634800550206327, | |
| "loss": 0.8732, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00026141678129298484, | |
| "loss": 0.8571, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00025935350756533696, | |
| "loss": 0.8597, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002572902338376891, | |
| "loss": 0.8719, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002552269601100413, | |
| "loss": 0.8624, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0002531636863823934, | |
| "loss": 0.8577, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00025110041265474553, | |
| "loss": 0.8624, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00024903713892709766, | |
| "loss": 0.8765, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0002469738651994498, | |
| "loss": 0.8595, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002449105914718019, | |
| "loss": 0.8646, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00024284731774415404, | |
| "loss": 0.87, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00024078404401650617, | |
| "loss": 0.8568, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002387207702888583, | |
| "loss": 0.8571, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_loss": 0.8650650382041931, | |
| "eval_runtime": 175.1777, | |
| "eval_samples_per_second": 11.417, | |
| "eval_steps_per_second": 0.285, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00023665749656121043, | |
| "loss": 0.8495, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00023459422283356255, | |
| "loss": 0.8572, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0002325309491059147, | |
| "loss": 0.8629, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00023046767537826684, | |
| "loss": 0.8413, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00022840440165061896, | |
| "loss": 0.853, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002263411279229711, | |
| "loss": 0.8662, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.00022427785419532322, | |
| "loss": 0.8495, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00022221458046767537, | |
| "loss": 0.8502, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002201513067400275, | |
| "loss": 0.8601, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00021808803301237963, | |
| "loss": 0.8503, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00021602475928473176, | |
| "loss": 0.8517, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00021396148555708389, | |
| "loss": 0.8633, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00021189821182943601, | |
| "loss": 0.853, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00020983493810178817, | |
| "loss": 0.8379, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0002077716643741403, | |
| "loss": 0.8396, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00020570839064649242, | |
| "loss": 0.8569, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00020364511691884455, | |
| "loss": 0.8594, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00020158184319119668, | |
| "loss": 0.8461, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00019951856946354884, | |
| "loss": 0.8429, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00019745529573590096, | |
| "loss": 0.8308, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.8592662215232849, | |
| "eval_runtime": 175.4031, | |
| "eval_samples_per_second": 11.402, | |
| "eval_steps_per_second": 0.285, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0001953920220082531, | |
| "loss": 0.8355, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0001933287482806052, | |
| "loss": 0.834, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00019126547455295732, | |
| "loss": 0.8299, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00018920220082530945, | |
| "loss": 0.8431, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00018713892709766163, | |
| "loss": 0.841, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00018507565337001373, | |
| "loss": 0.8374, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00018301237964236586, | |
| "loss": 0.84, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.000180949105914718, | |
| "loss": 0.8365, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00017888583218707011, | |
| "loss": 0.8287, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00017682255845942227, | |
| "loss": 0.8334, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0001747592847317744, | |
| "loss": 0.8445, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00017269601100412653, | |
| "loss": 0.8286, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00017063273727647865, | |
| "loss": 0.8266, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00016856946354883078, | |
| "loss": 0.8276, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00016650618982118294, | |
| "loss": 0.8386, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00016444291609353506, | |
| "loss": 0.8349, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0001623796423658872, | |
| "loss": 0.8292, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00016031636863823932, | |
| "loss": 0.8354, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00015825309491059145, | |
| "loss": 0.8306, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00015618982118294358, | |
| "loss": 0.8346, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_loss": 0.856368899345398, | |
| "eval_runtime": 175.5494, | |
| "eval_samples_per_second": 11.393, | |
| "eval_steps_per_second": 0.285, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00015412654745529573, | |
| "loss": 0.8244, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00015206327372764786, | |
| "loss": 0.8129, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00015, | |
| "loss": 0.8216, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00014793672627235211, | |
| "loss": 0.8205, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014587345254470424, | |
| "loss": 0.8215, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00014381017881705637, | |
| "loss": 0.8467, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0001417469050894085, | |
| "loss": 0.8258, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00013968363136176065, | |
| "loss": 0.8277, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00013762035763411278, | |
| "loss": 0.8249, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.0001355570839064649, | |
| "loss": 0.8226, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00013349381017881704, | |
| "loss": 0.8268, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00013143053645116917, | |
| "loss": 0.8321, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00012936726272352132, | |
| "loss": 0.8163, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00012730398899587345, | |
| "loss": 0.8352, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00012524071526822558, | |
| "loss": 0.8106, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001231774415405777, | |
| "loss": 0.8017, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00012111416781292983, | |
| "loss": 0.8244, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00011905089408528197, | |
| "loss": 0.8363, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0001169876203576341, | |
| "loss": 0.8307, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00011492434662998623, | |
| "loss": 0.8239, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.8538553714752197, | |
| "eval_runtime": 175.4751, | |
| "eval_samples_per_second": 11.398, | |
| "eval_steps_per_second": 0.285, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.00011286107290233837, | |
| "loss": 0.8269, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0001107977991746905, | |
| "loss": 0.8238, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010873452544704263, | |
| "loss": 0.8175, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.00010667125171939477, | |
| "loss": 0.8139, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.0001046079779917469, | |
| "loss": 0.8025, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00010254470426409902, | |
| "loss": 0.8056, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00010048143053645117, | |
| "loss": 0.8111, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.841815680880329e-05, | |
| "loss": 0.8163, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.635488308115543e-05, | |
| "loss": 0.8198, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.429160935350756e-05, | |
| "loss": 0.8019, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.222833562585969e-05, | |
| "loss": 0.8032, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.016506189821183e-05, | |
| "loss": 0.8105, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.810178817056396e-05, | |
| "loss": 0.8038, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.603851444291607e-05, | |
| "loss": 0.8089, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.397524071526822e-05, | |
| "loss": 0.7989, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.191196698762034e-05, | |
| "loss": 0.8091, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.984869325997248e-05, | |
| "loss": 0.8178, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 7.778541953232461e-05, | |
| "loss": 0.8027, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.572214580467674e-05, | |
| "loss": 0.8074, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 7.365887207702888e-05, | |
| "loss": 0.8106, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_loss": 0.8536396622657776, | |
| "eval_runtime": 175.6344, | |
| "eval_samples_per_second": 11.387, | |
| "eval_steps_per_second": 0.285, | |
| "step": 1200 | |
| } | |
| ], | |
| "max_steps": 1554, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.0681954887765328e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |