| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0631578947368423, | |
| "eval_steps": 25, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03368421052631579, | |
| "grad_norm": 0.39811971783638, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1869, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03368421052631579, | |
| "eval_loss": 2.1871018409729004, | |
| "eval_runtime": 51.726, | |
| "eval_samples_per_second": 1.933, | |
| "eval_steps_per_second": 0.967, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06736842105263158, | |
| "grad_norm": 0.43366214632987976, | |
| "learning_rate": 2e-05, | |
| "loss": 2.124, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10105263157894737, | |
| "grad_norm": 0.41468173265457153, | |
| "learning_rate": 3e-05, | |
| "loss": 2.1397, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13473684210526315, | |
| "grad_norm": 0.41099846363067627, | |
| "learning_rate": 4e-05, | |
| "loss": 2.1909, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16842105263157894, | |
| "grad_norm": 0.35164669156074524, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1483, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20210526315789473, | |
| "grad_norm": 0.3426206111907959, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0835, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23578947368421052, | |
| "grad_norm": 0.30055052042007446, | |
| "learning_rate": 7e-05, | |
| "loss": 2.1794, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2694736842105263, | |
| "grad_norm": 0.3416151702404022, | |
| "learning_rate": 8e-05, | |
| "loss": 2.1128, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3031578947368421, | |
| "grad_norm": 0.3587937355041504, | |
| "learning_rate": 9e-05, | |
| "loss": 2.1879, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3368421052631579, | |
| "grad_norm": 0.3512348234653473, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1369, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3705263157894737, | |
| "grad_norm": 0.3542545437812805, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 2.0747, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.40421052631578946, | |
| "grad_norm": 0.3271574079990387, | |
| "learning_rate": 0.00012, | |
| "loss": 2.0892, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4378947368421053, | |
| "grad_norm": 0.29716193675994873, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 2.0945, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47157894736842104, | |
| "grad_norm": 0.2834334373474121, | |
| "learning_rate": 0.00014, | |
| "loss": 2.1163, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5052631578947369, | |
| "grad_norm": 0.2676835060119629, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.0774, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5389473684210526, | |
| "grad_norm": 0.3038509786128998, | |
| "learning_rate": 0.00016, | |
| "loss": 2.0738, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5726315789473684, | |
| "grad_norm": 0.30092963576316833, | |
| "learning_rate": 0.00017, | |
| "loss": 2.0443, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6063157894736843, | |
| "grad_norm": 0.2652921974658966, | |
| "learning_rate": 0.00018, | |
| "loss": 2.0226, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.23126669228076935, | |
| "learning_rate": 0.00019, | |
| "loss": 2.0893, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6736842105263158, | |
| "grad_norm": 0.25947996973991394, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0797, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7073684210526315, | |
| "grad_norm": 0.24937580525875092, | |
| "learning_rate": 0.00019989930665413147, | |
| "loss": 2.113, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7410526315789474, | |
| "grad_norm": 0.23798011243343353, | |
| "learning_rate": 0.00019959742939952392, | |
| "loss": 2.0782, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.7747368421052632, | |
| "grad_norm": 0.23232997953891754, | |
| "learning_rate": 0.00019909497617679348, | |
| "loss": 2.0749, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8084210526315789, | |
| "grad_norm": 0.22174346446990967, | |
| "learning_rate": 0.00019839295885986296, | |
| "loss": 2.0835, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 0.20978115499019623, | |
| "learning_rate": 0.00019749279121818235, | |
| "loss": 2.1273, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "eval_loss": 2.085695743560791, | |
| "eval_runtime": 51.638, | |
| "eval_samples_per_second": 1.937, | |
| "eval_steps_per_second": 0.968, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8757894736842106, | |
| "grad_norm": 0.20944544672966003, | |
| "learning_rate": 0.00019639628606958533, | |
| "loss": 2.0262, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.9094736842105263, | |
| "grad_norm": 0.20344075560569763, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 2.0426, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9431578947368421, | |
| "grad_norm": 0.22783611714839935, | |
| "learning_rate": 0.00019362348706397373, | |
| "loss": 2.0554, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9768421052631578, | |
| "grad_norm": 0.19957782328128815, | |
| "learning_rate": 0.0001919527772551451, | |
| "loss": 2.0124, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0210526315789474, | |
| "grad_norm": 0.38232752680778503, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 3.3482, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0547368421052632, | |
| "grad_norm": 0.21452440321445465, | |
| "learning_rate": 0.0001880595531856738, | |
| "loss": 1.9092, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.088421052631579, | |
| "grad_norm": 0.20885184407234192, | |
| "learning_rate": 0.00018584487936018661, | |
| "loss": 2.0082, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.1221052631578947, | |
| "grad_norm": 0.23769386112689972, | |
| "learning_rate": 0.00018345732537213027, | |
| "loss": 1.9497, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1557894736842105, | |
| "grad_norm": 0.22073900699615479, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 2.0897, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1894736842105262, | |
| "grad_norm": 0.20628167688846588, | |
| "learning_rate": 0.000178183148246803, | |
| "loss": 1.9091, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.223157894736842, | |
| "grad_norm": 0.20254100859165192, | |
| "learning_rate": 0.00017530714660036112, | |
| "loss": 2.0089, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.256842105263158, | |
| "grad_norm": 0.21576657891273499, | |
| "learning_rate": 0.00017227948638273916, | |
| "loss": 1.9733, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2905263157894737, | |
| "grad_norm": 0.21345406770706177, | |
| "learning_rate": 0.00016910626489868649, | |
| "loss": 1.9593, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.3242105263157895, | |
| "grad_norm": 0.2377728521823883, | |
| "learning_rate": 0.00016579387259397127, | |
| "loss": 1.9921, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.3578947368421053, | |
| "grad_norm": 0.22072383761405945, | |
| "learning_rate": 0.00016234898018587337, | |
| "loss": 1.9538, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.391578947368421, | |
| "grad_norm": 0.22244809567928314, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 1.9346, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.4252631578947368, | |
| "grad_norm": 0.22625932097434998, | |
| "learning_rate": 0.00015508969814521025, | |
| "loss": 2.0541, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4589473684210525, | |
| "grad_norm": 0.22077956795692444, | |
| "learning_rate": 0.00015128992774059063, | |
| "loss": 1.9145, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.4926315789473685, | |
| "grad_norm": 0.22373822331428528, | |
| "learning_rate": 0.00014738686624729986, | |
| "loss": 2.0576, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.526315789473684, | |
| "grad_norm": 0.271822988986969, | |
| "learning_rate": 0.00014338837391175582, | |
| "loss": 1.9241, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.22692303359508514, | |
| "learning_rate": 0.00013930250316539238, | |
| "loss": 2.0193, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.5936842105263158, | |
| "grad_norm": 0.23916593194007874, | |
| "learning_rate": 0.0001351374824081343, | |
| "loss": 1.9436, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.6273684210526316, | |
| "grad_norm": 0.2340150773525238, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 1.9656, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6610526315789473, | |
| "grad_norm": 0.23769573867321014, | |
| "learning_rate": 0.00012660368455666752, | |
| "loss": 1.9992, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.694736842105263, | |
| "grad_norm": 0.227084219455719, | |
| "learning_rate": 0.00012225209339563145, | |
| "loss": 2.0096, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.694736842105263, | |
| "eval_loss": 2.0874733924865723, | |
| "eval_runtime": 51.8934, | |
| "eval_samples_per_second": 1.927, | |
| "eval_steps_per_second": 0.964, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.728421052631579, | |
| "grad_norm": 0.2218991219997406, | |
| "learning_rate": 0.00011785568947986367, | |
| "loss": 1.9305, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.7621052631578946, | |
| "grad_norm": 0.2461809664964676, | |
| "learning_rate": 0.00011342332658176555, | |
| "loss": 1.9155, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.7957894736842106, | |
| "grad_norm": 0.246916726231575, | |
| "learning_rate": 0.00010896393089034336, | |
| "loss": 1.9407, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.8294736842105264, | |
| "grad_norm": 0.23680198192596436, | |
| "learning_rate": 0.00010448648303505151, | |
| "loss": 1.8634, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.8631578947368421, | |
| "grad_norm": 0.2648875117301941, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0646, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.8968421052631579, | |
| "grad_norm": 0.24859718978405, | |
| "learning_rate": 9.551351696494854e-05, | |
| "loss": 1.9494, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9305263157894736, | |
| "grad_norm": 0.26295167207717896, | |
| "learning_rate": 9.103606910965666e-05, | |
| "loss": 1.9067, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.9642105263157896, | |
| "grad_norm": 0.24121926724910736, | |
| "learning_rate": 8.657667341823448e-05, | |
| "loss": 1.9897, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.008421052631579, | |
| "grad_norm": 0.4270052909851074, | |
| "learning_rate": 8.214431052013634e-05, | |
| "loss": 3.2422, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.042105263157895, | |
| "grad_norm": 0.2644731402397156, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 1.8985, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0757894736842104, | |
| "grad_norm": 0.26980847120285034, | |
| "learning_rate": 7.339631544333249e-05, | |
| "loss": 1.8777, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.1094736842105264, | |
| "grad_norm": 0.25031837821006775, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 1.8957, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.143157894736842, | |
| "grad_norm": 0.2601235806941986, | |
| "learning_rate": 6.486251759186572e-05, | |
| "loss": 1.8774, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.176842105263158, | |
| "grad_norm": 0.27925577759742737, | |
| "learning_rate": 6.069749683460765e-05, | |
| "loss": 1.8837, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.2105263157894735, | |
| "grad_norm": 0.2980926036834717, | |
| "learning_rate": 5.6611626088244194e-05, | |
| "loss": 1.8308, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2442105263157894, | |
| "grad_norm": 0.3060623109340668, | |
| "learning_rate": 5.261313375270014e-05, | |
| "loss": 1.8415, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.2778947368421054, | |
| "grad_norm": 0.2616610825061798, | |
| "learning_rate": 4.87100722594094e-05, | |
| "loss": 1.8877, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.311578947368421, | |
| "grad_norm": 0.26330044865608215, | |
| "learning_rate": 4.491030185478976e-05, | |
| "loss": 1.9152, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.345263157894737, | |
| "grad_norm": 0.2531837224960327, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 1.753, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.3789473684210525, | |
| "grad_norm": 0.29288211464881897, | |
| "learning_rate": 3.7651019814126654e-05, | |
| "loss": 1.8325, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.4126315789473685, | |
| "grad_norm": 0.2796313762664795, | |
| "learning_rate": 3.4206127406028745e-05, | |
| "loss": 1.8628, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.446315789473684, | |
| "grad_norm": 0.2576499283313751, | |
| "learning_rate": 3.089373510131354e-05, | |
| "loss": 1.8157, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.26932939887046814, | |
| "learning_rate": 2.7720513617260856e-05, | |
| "loss": 1.7607, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.513684210526316, | |
| "grad_norm": 0.2623027563095093, | |
| "learning_rate": 2.4692853399638917e-05, | |
| "loss": 1.8798, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.5473684210526315, | |
| "grad_norm": 0.2754516005516052, | |
| "learning_rate": 2.181685175319702e-05, | |
| "loss": 1.8257, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5473684210526315, | |
| "eval_loss": 2.1104278564453125, | |
| "eval_runtime": 51.5518, | |
| "eval_samples_per_second": 1.94, | |
| "eval_steps_per_second": 0.97, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5810526315789475, | |
| "grad_norm": 0.2684014141559601, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.8146, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.614736842105263, | |
| "grad_norm": 0.27565282583236694, | |
| "learning_rate": 1.6542674627869737e-05, | |
| "loss": 1.9054, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.648421052631579, | |
| "grad_norm": 0.27578240633010864, | |
| "learning_rate": 1.415512063981339e-05, | |
| "loss": 1.8384, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.6821052631578945, | |
| "grad_norm": 0.2741873264312744, | |
| "learning_rate": 1.19404468143262e-05, | |
| "loss": 1.7925, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.7157894736842105, | |
| "grad_norm": 0.28904208540916443, | |
| "learning_rate": 9.903113209758096e-06, | |
| "loss": 1.8668, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.7494736842105265, | |
| "grad_norm": 0.291592538356781, | |
| "learning_rate": 8.047222744854943e-06, | |
| "loss": 1.8642, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.783157894736842, | |
| "grad_norm": 0.287810742855072, | |
| "learning_rate": 6.37651293602628e-06, | |
| "loss": 1.8866, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.816842105263158, | |
| "grad_norm": 0.28119951486587524, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 1.7627, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.8505263157894736, | |
| "grad_norm": 0.26779597997665405, | |
| "learning_rate": 3.6037139304146762e-06, | |
| "loss": 1.9118, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.8842105263157896, | |
| "grad_norm": 0.2809458374977112, | |
| "learning_rate": 2.5072087818176382e-06, | |
| "loss": 1.9202, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.917894736842105, | |
| "grad_norm": 0.27657240629196167, | |
| "learning_rate": 1.6070411401370334e-06, | |
| "loss": 1.7985, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.951578947368421, | |
| "grad_norm": 0.2845059335231781, | |
| "learning_rate": 9.0502382320653e-07, | |
| "loss": 1.7862, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.985263157894737, | |
| "grad_norm": 0.4435933828353882, | |
| "learning_rate": 4.025706004760932e-07, | |
| "loss": 2.6286, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.0294736842105263, | |
| "grad_norm": 0.3721281886100769, | |
| "learning_rate": 1.0069334586854107e-07, | |
| "loss": 2.1659, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.0631578947368423, | |
| "grad_norm": 0.2612936794757843, | |
| "learning_rate": 0.0, | |
| "loss": 1.8654, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0742550511170355e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |