| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2578875171467764, | |
| "eval_steps": 500, | |
| "global_step": 94, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027434842249657062, | |
| "grad_norm": 0.3216973543167114, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.039, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0054869684499314125, | |
| "grad_norm": 0.3218615651130676, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.063, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00823045267489712, | |
| "grad_norm": 0.3071180582046509, | |
| "learning_rate": 6e-06, | |
| "loss": 2.0298, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010973936899862825, | |
| "grad_norm": 0.30708086490631104, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.9489, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013717421124828532, | |
| "grad_norm": 0.31064078211784363, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0401, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01646090534979424, | |
| "grad_norm": 0.35925883054733276, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.0963, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.019204389574759947, | |
| "grad_norm": 0.3267571032047272, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 2.0252, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02194787379972565, | |
| "grad_norm": 0.30728623270988464, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.0108, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "grad_norm": 0.2928607761859894, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.9542, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027434842249657063, | |
| "grad_norm": 0.30577352643013, | |
| "learning_rate": 2e-05, | |
| "loss": 2.017, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03017832647462277, | |
| "grad_norm": 0.3024803102016449, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.982, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03292181069958848, | |
| "grad_norm": 0.28839072585105896, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.9217, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03566529492455418, | |
| "grad_norm": 0.2843893766403198, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.9608, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.038408779149519894, | |
| "grad_norm": 0.2703002095222473, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.9696, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0411522633744856, | |
| "grad_norm": 0.24636265635490417, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8818, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0438957475994513, | |
| "grad_norm": 0.2405432015657425, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.975, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04663923182441701, | |
| "grad_norm": 0.24582137167453766, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.923, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04938271604938271, | |
| "grad_norm": 0.2504767179489136, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.9781, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05212620027434842, | |
| "grad_norm": 0.2394665777683258, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.8959, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05486968449931413, | |
| "grad_norm": 0.24969030916690826, | |
| "learning_rate": 4e-05, | |
| "loss": 1.855, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05761316872427984, | |
| "grad_norm": 0.2694351077079773, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.9602, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.06035665294924554, | |
| "grad_norm": 0.25622957944869995, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.8208, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06310013717421124, | |
| "grad_norm": 0.24535588920116425, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.7967, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06584362139917696, | |
| "grad_norm": 0.2737885117530823, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.7841, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06858710562414266, | |
| "grad_norm": 0.2646300196647644, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7744, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.07133058984910837, | |
| "grad_norm": 0.2676407992839813, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 1.7359, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 0.2649776041507721, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 1.7205, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07681755829903979, | |
| "grad_norm": 0.296818345785141, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.696, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07956104252400549, | |
| "grad_norm": 0.31905728578567505, | |
| "learning_rate": 5.8e-05, | |
| "loss": 1.7261, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0823045267489712, | |
| "grad_norm": 0.4174517095088959, | |
| "learning_rate": 6e-05, | |
| "loss": 1.6451, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0850480109739369, | |
| "grad_norm": 0.4545894265174866, | |
| "learning_rate": 6.2e-05, | |
| "loss": 1.5867, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0877914951989026, | |
| "grad_norm": 0.45722702145576477, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.5184, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09053497942386832, | |
| "grad_norm": 0.4953472316265106, | |
| "learning_rate": 6.6e-05, | |
| "loss": 1.4793, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.09327846364883402, | |
| "grad_norm": 0.5516601800918579, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.4967, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09602194787379972, | |
| "grad_norm": 0.5295405983924866, | |
| "learning_rate": 7e-05, | |
| "loss": 1.4445, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09876543209876543, | |
| "grad_norm": 0.3918333351612091, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.3956, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10150891632373114, | |
| "grad_norm": 0.4032560884952545, | |
| "learning_rate": 7.4e-05, | |
| "loss": 1.3773, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.10425240054869685, | |
| "grad_norm": 0.30622419714927673, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.2721, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10699588477366255, | |
| "grad_norm": 0.2740858495235443, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.3173, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10973936899862825, | |
| "grad_norm": 0.19876384735107422, | |
| "learning_rate": 8e-05, | |
| "loss": 1.388, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11248285322359397, | |
| "grad_norm": 0.20184342563152313, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.2856, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.11522633744855967, | |
| "grad_norm": 0.1708114743232727, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.3209, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11796982167352538, | |
| "grad_norm": 0.1288367211818695, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.2782, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.12071330589849108, | |
| "grad_norm": 0.12842676043510437, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.3378, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12345679012345678, | |
| "grad_norm": 0.12385573983192444, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2032, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1262002743484225, | |
| "grad_norm": 0.12317045032978058, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.3391, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1289437585733882, | |
| "grad_norm": 0.126034677028656, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.2907, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.13168724279835392, | |
| "grad_norm": 0.1402864307165146, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2058, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13443072702331962, | |
| "grad_norm": 0.13603994250297546, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.2651, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.13717421124828533, | |
| "grad_norm": 0.13384652137756348, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3055, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13991769547325103, | |
| "grad_norm": 0.12770739197731018, | |
| "learning_rate": 0.00010200000000000001, | |
| "loss": 1.2601, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.14266117969821673, | |
| "grad_norm": 0.13551487028598785, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 1.2371, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14540466392318244, | |
| "grad_norm": 0.14969402551651, | |
| "learning_rate": 0.00010600000000000002, | |
| "loss": 1.1847, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 0.15096262097358704, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 1.2237, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.15089163237311384, | |
| "grad_norm": 0.167573019862175, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 1.268, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.15363511659807957, | |
| "grad_norm": 0.16625213623046875, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 1.23, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15637860082304528, | |
| "grad_norm": 0.17408426105976105, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 1.2036, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.15912208504801098, | |
| "grad_norm": 0.19839078187942505, | |
| "learning_rate": 0.000116, | |
| "loss": 1.1494, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.16186556927297668, | |
| "grad_norm": 0.1805109679698944, | |
| "learning_rate": 0.000118, | |
| "loss": 1.2177, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.1646090534979424, | |
| "grad_norm": 0.19576989114284515, | |
| "learning_rate": 0.00012, | |
| "loss": 1.2494, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1673525377229081, | |
| "grad_norm": 0.21191267669200897, | |
| "learning_rate": 0.000122, | |
| "loss": 1.2526, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.1700960219478738, | |
| "grad_norm": 0.20439420640468597, | |
| "learning_rate": 0.000124, | |
| "loss": 1.193, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.1728395061728395, | |
| "grad_norm": 0.1174619197845459, | |
| "learning_rate": 0.000126, | |
| "loss": 1.1726, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.1755829903978052, | |
| "grad_norm": 0.21233001351356506, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 1.1495, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17832647462277093, | |
| "grad_norm": 0.20875731110572815, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 1.1956, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.18106995884773663, | |
| "grad_norm": 0.17120110988616943, | |
| "learning_rate": 0.000132, | |
| "loss": 1.1818, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.18381344307270234, | |
| "grad_norm": 0.1390346735715866, | |
| "learning_rate": 0.000134, | |
| "loss": 1.1771, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.18655692729766804, | |
| "grad_norm": 0.09387281537055969, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 1.1983, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18930041152263374, | |
| "grad_norm": 0.07457795739173889, | |
| "learning_rate": 0.000138, | |
| "loss": 1.1385, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.19204389574759945, | |
| "grad_norm": 0.08271007239818573, | |
| "learning_rate": 0.00014, | |
| "loss": 1.127, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.19478737997256515, | |
| "grad_norm": 0.07676747441291809, | |
| "learning_rate": 0.000142, | |
| "loss": 1.1444, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.0707523301243782, | |
| "learning_rate": 0.000144, | |
| "loss": 1.1589, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.20027434842249658, | |
| "grad_norm": 0.0701480582356453, | |
| "learning_rate": 0.000146, | |
| "loss": 1.1631, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.2030178326474623, | |
| "grad_norm": 0.07315018028020859, | |
| "learning_rate": 0.000148, | |
| "loss": 1.1915, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.205761316872428, | |
| "grad_norm": 0.0682872086763382, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.2017, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2085048010973937, | |
| "grad_norm": 0.07075867056846619, | |
| "learning_rate": 0.000152, | |
| "loss": 1.1562, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2112482853223594, | |
| "grad_norm": 0.06364033371210098, | |
| "learning_rate": 0.000154, | |
| "loss": 1.1936, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.2139917695473251, | |
| "grad_norm": 0.06413716077804565, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 1.1463, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.2167352537722908, | |
| "grad_norm": 0.06316008418798447, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 1.1975, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.2194787379972565, | |
| "grad_norm": 0.06660479307174683, | |
| "learning_rate": 0.00016, | |
| "loss": 1.1684, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.06477335095405579, | |
| "learning_rate": 0.000162, | |
| "loss": 1.1851, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.22496570644718794, | |
| "grad_norm": 0.0677405372262001, | |
| "learning_rate": 0.000164, | |
| "loss": 1.093, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.22770919067215364, | |
| "grad_norm": 0.06988447159528732, | |
| "learning_rate": 0.000166, | |
| "loss": 1.2542, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.23045267489711935, | |
| "grad_norm": 0.06364695727825165, | |
| "learning_rate": 0.000168, | |
| "loss": 1.0451, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.23319615912208505, | |
| "grad_norm": 0.06621105968952179, | |
| "learning_rate": 0.00017, | |
| "loss": 1.1022, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.23593964334705075, | |
| "grad_norm": 0.06702058762311935, | |
| "learning_rate": 0.000172, | |
| "loss": 1.1803, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.23868312757201646, | |
| "grad_norm": 0.06412065774202347, | |
| "learning_rate": 0.000174, | |
| "loss": 1.1867, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.24142661179698216, | |
| "grad_norm": 0.06154424324631691, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 1.2207, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.24417009602194786, | |
| "grad_norm": 0.06396840512752533, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 1.1466, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.24691358024691357, | |
| "grad_norm": 0.06146432086825371, | |
| "learning_rate": 0.00018, | |
| "loss": 1.1748, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2496570644718793, | |
| "grad_norm": 0.06494217365980148, | |
| "learning_rate": 0.000182, | |
| "loss": 1.1481, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.252400548696845, | |
| "grad_norm": 0.06837109476327896, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 1.2352, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2551440329218107, | |
| "grad_norm": 0.06343371421098709, | |
| "learning_rate": 0.00018600000000000002, | |
| "loss": 1.1362, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.2578875171467764, | |
| "grad_norm": 0.06184321269392967, | |
| "learning_rate": 0.000188, | |
| "loss": 1.1256, | |
| "step": 94 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 364, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.6971271019115315e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |