diff --git "a/checkpoint-738/trainer_state.json" "b/checkpoint-738/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-738/trainer_state.json" @@ -0,0 +1,5829 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9077490774907749, + "eval_steps": 21, + "global_step": 738, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012300123001230013, + "grad_norm": 6.540346145629883, + "learning_rate": 9.803921568627452e-08, + "loss": 0.3208, + "step": 1 + }, + { + "epoch": 0.0024600246002460025, + "grad_norm": 5.055933475494385, + "learning_rate": 1.9607843137254904e-07, + "loss": 0.1703, + "step": 2 + }, + { + "epoch": 0.0036900369003690036, + "grad_norm": 6.361550331115723, + "learning_rate": 2.9411764705882356e-07, + "loss": 0.3362, + "step": 3 + }, + { + "epoch": 0.004920049200492005, + "grad_norm": 6.709433078765869, + "learning_rate": 3.921568627450981e-07, + "loss": 0.3346, + "step": 4 + }, + { + "epoch": 0.006150061500615006, + "grad_norm": 5.4415154457092285, + "learning_rate": 4.901960784313725e-07, + "loss": 0.2484, + "step": 5 + }, + { + "epoch": 0.007380073800738007, + "grad_norm": 5.709558010101318, + "learning_rate": 5.882352941176471e-07, + "loss": 0.2249, + "step": 6 + }, + { + "epoch": 0.008610086100861008, + "grad_norm": 6.553178787231445, + "learning_rate": 6.862745098039217e-07, + "loss": 0.2724, + "step": 7 + }, + { + "epoch": 0.00984009840098401, + "grad_norm": 5.640111446380615, + "learning_rate": 7.843137254901962e-07, + "loss": 0.251, + "step": 8 + }, + { + "epoch": 0.01107011070110701, + "grad_norm": 5.696380615234375, + "learning_rate": 8.823529411764707e-07, + "loss": 0.2413, + "step": 9 + }, + { + "epoch": 0.012300123001230012, + "grad_norm": 6.983877182006836, + "learning_rate": 9.80392156862745e-07, + "loss": 0.382, + "step": 10 + }, + { + "epoch": 0.013530135301353014, + "grad_norm": 6.066723346710205, + "learning_rate": 1.0784313725490197e-06, + "loss": 0.2695, + "step": 11 + }, + { + "epoch": 0.014760147601476014, + "grad_norm": 5.643115520477295, + "learning_rate": 1.1764705882352942e-06, + "loss": 0.2392, + "step": 12 + }, + { + "epoch": 0.015990159901599015, + "grad_norm": 6.062892436981201, + "learning_rate": 1.2745098039215686e-06, + "loss": 0.3603, + "step": 13 + }, + { + "epoch": 0.017220172201722016, + "grad_norm": 6.2491655349731445, + "learning_rate": 1.3725490196078434e-06, + "loss": 0.3282, + "step": 14 + }, + { + "epoch": 0.01845018450184502, + "grad_norm": 6.1164398193359375, + "learning_rate": 1.4705882352941177e-06, + "loss": 0.2878, + "step": 15 + }, + { + "epoch": 0.01968019680196802, + "grad_norm": 5.676611423492432, + "learning_rate": 1.5686274509803923e-06, + "loss": 0.3046, + "step": 16 + }, + { + "epoch": 0.020910209102091022, + "grad_norm": 7.181272983551025, + "learning_rate": 1.6666666666666667e-06, + "loss": 0.3946, + "step": 17 + }, + { + "epoch": 0.02214022140221402, + "grad_norm": 5.430984020233154, + "learning_rate": 1.7647058823529414e-06, + "loss": 0.2038, + "step": 18 + }, + { + "epoch": 0.023370233702337023, + "grad_norm": 7.2283220291137695, + "learning_rate": 1.8627450980392158e-06, + "loss": 0.3542, + "step": 19 + }, + { + "epoch": 0.024600246002460024, + "grad_norm": 5.587338924407959, + "learning_rate": 1.96078431372549e-06, + "loss": 0.2369, + "step": 20 + }, + { + "epoch": 0.025830258302583026, + "grad_norm": 4.456090927124023, + "learning_rate": 2.058823529411765e-06, + "loss": 0.1967, + "step": 21 + }, + { + "epoch": 0.025830258302583026, + "eval_loss": 0.14506277441978455, + "eval_runtime": 54.872, + "eval_samples_per_second": 27.446, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8860152816653839, + "eval_sts-test_pearson_dot": 0.8766503125978379, + "eval_sts-test_pearson_euclidean": 0.9084101290541164, + "eval_sts-test_pearson_manhattan": 0.909121525028934, + "eval_sts-test_pearson_max": 0.909121525028934, + "eval_sts-test_spearman_cosine": 0.9080919696366193, + "eval_sts-test_spearman_dot": 0.8799434709726907, + "eval_sts-test_spearman_euclidean": 0.9044399981995129, + "eval_sts-test_spearman_manhattan": 0.9048055712538192, + "eval_sts-test_spearman_max": 0.9080919696366193, + "step": 21 + }, + { + "epoch": 0.02706027060270603, + "grad_norm": 6.088884353637695, + "learning_rate": 2.1568627450980393e-06, + "loss": 0.2368, + "step": 22 + }, + { + "epoch": 0.028290282902829027, + "grad_norm": 5.354013919830322, + "learning_rate": 2.254901960784314e-06, + "loss": 0.263, + "step": 23 + }, + { + "epoch": 0.02952029520295203, + "grad_norm": 7.822023391723633, + "learning_rate": 2.3529411764705885e-06, + "loss": 0.3595, + "step": 24 + }, + { + "epoch": 0.03075030750307503, + "grad_norm": 6.401333332061768, + "learning_rate": 2.450980392156863e-06, + "loss": 0.3073, + "step": 25 + }, + { + "epoch": 0.03198031980319803, + "grad_norm": 5.567343235015869, + "learning_rate": 2.549019607843137e-06, + "loss": 0.2232, + "step": 26 + }, + { + "epoch": 0.033210332103321034, + "grad_norm": 4.244979381561279, + "learning_rate": 2.647058823529412e-06, + "loss": 0.1822, + "step": 27 + }, + { + "epoch": 0.03444034440344403, + "grad_norm": 5.674376964569092, + "learning_rate": 2.7450980392156867e-06, + "loss": 0.251, + "step": 28 + }, + { + "epoch": 0.03567035670356704, + "grad_norm": 6.017494201660156, + "learning_rate": 2.843137254901961e-06, + "loss": 0.2677, + "step": 29 + }, + { + "epoch": 0.03690036900369004, + "grad_norm": 6.415028095245361, + "learning_rate": 2.9411764705882355e-06, + "loss": 0.3252, + "step": 30 + }, + { + "epoch": 0.038130381303813035, + "grad_norm": 5.484204292297363, + "learning_rate": 3.03921568627451e-06, + "loss": 0.2058, + "step": 31 + }, + { + "epoch": 0.03936039360393604, + "grad_norm": 5.997295379638672, + "learning_rate": 3.1372549019607846e-06, + "loss": 0.3083, + "step": 32 + }, + { + "epoch": 0.04059040590405904, + "grad_norm": 5.527047157287598, + "learning_rate": 3.2352941176470594e-06, + "loss": 0.2109, + "step": 33 + }, + { + "epoch": 0.041820418204182044, + "grad_norm": 5.817302227020264, + "learning_rate": 3.3333333333333333e-06, + "loss": 0.2751, + "step": 34 + }, + { + "epoch": 0.04305043050430504, + "grad_norm": 5.476433753967285, + "learning_rate": 3.431372549019608e-06, + "loss": 0.2269, + "step": 35 + }, + { + "epoch": 0.04428044280442804, + "grad_norm": 5.363610744476318, + "learning_rate": 3.529411764705883e-06, + "loss": 0.2333, + "step": 36 + }, + { + "epoch": 0.04551045510455105, + "grad_norm": 6.07395601272583, + "learning_rate": 3.6274509803921573e-06, + "loss": 0.2747, + "step": 37 + }, + { + "epoch": 0.046740467404674045, + "grad_norm": 4.726163864135742, + "learning_rate": 3.7254901960784316e-06, + "loss": 0.1285, + "step": 38 + }, + { + "epoch": 0.04797047970479705, + "grad_norm": 5.783392906188965, + "learning_rate": 3.8235294117647055e-06, + "loss": 0.3659, + "step": 39 + }, + { + "epoch": 0.04920049200492005, + "grad_norm": 6.566931247711182, + "learning_rate": 3.92156862745098e-06, + "loss": 0.3991, + "step": 40 + }, + { + "epoch": 0.05043050430504305, + "grad_norm": 5.311452388763428, + "learning_rate": 4.019607843137255e-06, + "loss": 0.2647, + "step": 41 + }, + { + "epoch": 0.05166051660516605, + "grad_norm": 6.0737152099609375, + "learning_rate": 4.11764705882353e-06, + "loss": 0.3627, + "step": 42 + }, + { + "epoch": 0.05166051660516605, + "eval_loss": 0.1373225301504135, + "eval_runtime": 54.8187, + "eval_samples_per_second": 27.472, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8846111050777101, + "eval_sts-test_pearson_dot": 0.8747554197498655, + "eval_sts-test_pearson_euclidean": 0.9089352149126115, + "eval_sts-test_pearson_manhattan": 0.9098483550214526, + "eval_sts-test_pearson_max": 0.9098483550214526, + "eval_sts-test_spearman_cosine": 0.9084485029361248, + "eval_sts-test_spearman_dot": 0.8796038088987298, + "eval_sts-test_spearman_euclidean": 0.9055790073044468, + "eval_sts-test_spearman_manhattan": 0.9063848432683216, + "eval_sts-test_spearman_max": 0.9084485029361248, + "step": 42 + }, + { + "epoch": 0.05289052890528905, + "grad_norm": 4.857839584350586, + "learning_rate": 4.215686274509805e-06, + "loss": 0.2026, + "step": 43 + }, + { + "epoch": 0.05412054120541206, + "grad_norm": 5.248873233795166, + "learning_rate": 4.313725490196079e-06, + "loss": 0.1923, + "step": 44 + }, + { + "epoch": 0.055350553505535055, + "grad_norm": 5.329862117767334, + "learning_rate": 4.411764705882353e-06, + "loss": 0.2369, + "step": 45 + }, + { + "epoch": 0.056580565805658053, + "grad_norm": 5.581146240234375, + "learning_rate": 4.509803921568628e-06, + "loss": 0.2268, + "step": 46 + }, + { + "epoch": 0.05781057810578106, + "grad_norm": 5.818411350250244, + "learning_rate": 4.607843137254902e-06, + "loss": 0.2975, + "step": 47 + }, + { + "epoch": 0.05904059040590406, + "grad_norm": 5.096602916717529, + "learning_rate": 4.705882352941177e-06, + "loss": 0.1922, + "step": 48 + }, + { + "epoch": 0.06027060270602706, + "grad_norm": 5.256355285644531, + "learning_rate": 4.803921568627452e-06, + "loss": 0.1906, + "step": 49 + }, + { + "epoch": 0.06150061500615006, + "grad_norm": 5.3927388191223145, + "learning_rate": 4.901960784313726e-06, + "loss": 0.2379, + "step": 50 + }, + { + "epoch": 0.06273062730627306, + "grad_norm": 6.2723846435546875, + "learning_rate": 5e-06, + "loss": 0.3796, + "step": 51 + }, + { + "epoch": 0.06396063960639606, + "grad_norm": 4.595238208770752, + "learning_rate": 5.098039215686274e-06, + "loss": 0.1821, + "step": 52 + }, + { + "epoch": 0.06519065190651907, + "grad_norm": 4.342020511627197, + "learning_rate": 5.19607843137255e-06, + "loss": 0.1257, + "step": 53 + }, + { + "epoch": 0.06642066420664207, + "grad_norm": 4.998225212097168, + "learning_rate": 5.294117647058824e-06, + "loss": 0.2368, + "step": 54 + }, + { + "epoch": 0.06765067650676507, + "grad_norm": 5.510946273803711, + "learning_rate": 5.392156862745098e-06, + "loss": 0.294, + "step": 55 + }, + { + "epoch": 0.06888068880688807, + "grad_norm": 4.788788318634033, + "learning_rate": 5.4901960784313735e-06, + "loss": 0.2594, + "step": 56 + }, + { + "epoch": 0.07011070110701106, + "grad_norm": 5.827020645141602, + "learning_rate": 5.588235294117647e-06, + "loss": 0.2972, + "step": 57 + }, + { + "epoch": 0.07134071340713408, + "grad_norm": 4.821737289428711, + "learning_rate": 5.686274509803922e-06, + "loss": 0.2297, + "step": 58 + }, + { + "epoch": 0.07257072570725707, + "grad_norm": 4.880247592926025, + "learning_rate": 5.784313725490197e-06, + "loss": 0.1487, + "step": 59 + }, + { + "epoch": 0.07380073800738007, + "grad_norm": 4.447835445404053, + "learning_rate": 5.882352941176471e-06, + "loss": 0.182, + "step": 60 + }, + { + "epoch": 0.07503075030750307, + "grad_norm": 5.5556640625, + "learning_rate": 5.980392156862746e-06, + "loss": 0.2516, + "step": 61 + }, + { + "epoch": 0.07626076260762607, + "grad_norm": 5.217922687530518, + "learning_rate": 6.07843137254902e-06, + "loss": 0.2809, + "step": 62 + }, + { + "epoch": 0.07749077490774908, + "grad_norm": 4.436608791351318, + "learning_rate": 6.176470588235295e-06, + "loss": 0.1371, + "step": 63 + }, + { + "epoch": 0.07749077490774908, + "eval_loss": 0.13080179691314697, + "eval_runtime": 54.9188, + "eval_samples_per_second": 27.422, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.882074513745531, + "eval_sts-test_pearson_dot": 0.8709046425878566, + "eval_sts-test_pearson_euclidean": 0.9081794284297221, + "eval_sts-test_pearson_manhattan": 0.9093974331692458, + "eval_sts-test_pearson_max": 0.9093974331692458, + "eval_sts-test_spearman_cosine": 0.9067824582257844, + "eval_sts-test_spearman_dot": 0.8757477717096785, + "eval_sts-test_spearman_euclidean": 0.9051085820447002, + "eval_sts-test_spearman_manhattan": 0.9064308923162935, + "eval_sts-test_spearman_max": 0.9067824582257844, + "step": 63 + }, + { + "epoch": 0.07872078720787208, + "grad_norm": 5.6947021484375, + "learning_rate": 6.274509803921569e-06, + "loss": 0.2149, + "step": 64 + }, + { + "epoch": 0.07995079950799508, + "grad_norm": 4.272282600402832, + "learning_rate": 6.372549019607843e-06, + "loss": 0.1806, + "step": 65 + }, + { + "epoch": 0.08118081180811808, + "grad_norm": 4.575979232788086, + "learning_rate": 6.470588235294119e-06, + "loss": 0.1458, + "step": 66 + }, + { + "epoch": 0.08241082410824108, + "grad_norm": 4.315216541290283, + "learning_rate": 6.568627450980393e-06, + "loss": 0.249, + "step": 67 + }, + { + "epoch": 0.08364083640836409, + "grad_norm": 5.67277193069458, + "learning_rate": 6.666666666666667e-06, + "loss": 0.2787, + "step": 68 + }, + { + "epoch": 0.08487084870848709, + "grad_norm": 5.964886665344238, + "learning_rate": 6.764705882352942e-06, + "loss": 0.288, + "step": 69 + }, + { + "epoch": 0.08610086100861009, + "grad_norm": 4.218502521514893, + "learning_rate": 6.862745098039216e-06, + "loss": 0.1461, + "step": 70 + }, + { + "epoch": 0.08733087330873308, + "grad_norm": 5.179543972015381, + "learning_rate": 6.96078431372549e-06, + "loss": 0.2304, + "step": 71 + }, + { + "epoch": 0.08856088560885608, + "grad_norm": 5.720668792724609, + "learning_rate": 7.058823529411766e-06, + "loss": 0.3505, + "step": 72 + }, + { + "epoch": 0.0897908979089791, + "grad_norm": 5.2965497970581055, + "learning_rate": 7.15686274509804e-06, + "loss": 0.2227, + "step": 73 + }, + { + "epoch": 0.0910209102091021, + "grad_norm": 4.685606956481934, + "learning_rate": 7.2549019607843145e-06, + "loss": 0.1746, + "step": 74 + }, + { + "epoch": 0.09225092250922509, + "grad_norm": 4.2930145263671875, + "learning_rate": 7.352941176470589e-06, + "loss": 0.1484, + "step": 75 + }, + { + "epoch": 0.09348093480934809, + "grad_norm": 3.764916181564331, + "learning_rate": 7.450980392156863e-06, + "loss": 0.1346, + "step": 76 + }, + { + "epoch": 0.09471094710947109, + "grad_norm": 5.033151626586914, + "learning_rate": 7.549019607843138e-06, + "loss": 0.2112, + "step": 77 + }, + { + "epoch": 0.0959409594095941, + "grad_norm": 5.817330837249756, + "learning_rate": 7.647058823529411e-06, + "loss": 0.3138, + "step": 78 + }, + { + "epoch": 0.0971709717097171, + "grad_norm": 6.147035121917725, + "learning_rate": 7.745098039215687e-06, + "loss": 0.2675, + "step": 79 + }, + { + "epoch": 0.0984009840098401, + "grad_norm": 5.131881237030029, + "learning_rate": 7.84313725490196e-06, + "loss": 0.2849, + "step": 80 + }, + { + "epoch": 0.0996309963099631, + "grad_norm": 4.2269368171691895, + "learning_rate": 7.941176470588236e-06, + "loss": 0.1719, + "step": 81 + }, + { + "epoch": 0.1008610086100861, + "grad_norm": 5.200590133666992, + "learning_rate": 8.03921568627451e-06, + "loss": 0.2749, + "step": 82 + }, + { + "epoch": 0.10209102091020911, + "grad_norm": 5.44044303894043, + "learning_rate": 8.137254901960784e-06, + "loss": 0.3097, + "step": 83 + }, + { + "epoch": 0.1033210332103321, + "grad_norm": 4.603049278259277, + "learning_rate": 8.23529411764706e-06, + "loss": 0.2068, + "step": 84 + }, + { + "epoch": 0.1033210332103321, + "eval_loss": 0.1260141134262085, + "eval_runtime": 54.8932, + "eval_samples_per_second": 27.435, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8775839612260851, + "eval_sts-test_pearson_dot": 0.8664914414909934, + "eval_sts-test_pearson_euclidean": 0.9054210798291935, + "eval_sts-test_pearson_manhattan": 0.9069843565115414, + "eval_sts-test_pearson_max": 0.9069843565115414, + "eval_sts-test_spearman_cosine": 0.9044597335057865, + "eval_sts-test_spearman_dot": 0.872940077569982, + "eval_sts-test_spearman_euclidean": 0.9027100934391671, + "eval_sts-test_spearman_manhattan": 0.904476380975024, + "eval_sts-test_spearman_max": 0.904476380975024, + "step": 84 + }, + { + "epoch": 0.1045510455104551, + "grad_norm": 4.813210964202881, + "learning_rate": 8.333333333333334e-06, + "loss": 0.22, + "step": 85 + }, + { + "epoch": 0.1057810578105781, + "grad_norm": 4.659386157989502, + "learning_rate": 8.43137254901961e-06, + "loss": 0.2977, + "step": 86 + }, + { + "epoch": 0.1070110701107011, + "grad_norm": 4.895315647125244, + "learning_rate": 8.529411764705883e-06, + "loss": 0.209, + "step": 87 + }, + { + "epoch": 0.10824108241082411, + "grad_norm": 5.339110851287842, + "learning_rate": 8.627450980392157e-06, + "loss": 0.2215, + "step": 88 + }, + { + "epoch": 0.10947109471094711, + "grad_norm": 4.615406036376953, + "learning_rate": 8.725490196078433e-06, + "loss": 0.1948, + "step": 89 + }, + { + "epoch": 0.11070110701107011, + "grad_norm": 5.0383734703063965, + "learning_rate": 8.823529411764707e-06, + "loss": 0.2084, + "step": 90 + }, + { + "epoch": 0.11193111931119311, + "grad_norm": 3.9511592388153076, + "learning_rate": 8.921568627450982e-06, + "loss": 0.1823, + "step": 91 + }, + { + "epoch": 0.11316113161131611, + "grad_norm": 5.13690710067749, + "learning_rate": 9.019607843137256e-06, + "loss": 0.255, + "step": 92 + }, + { + "epoch": 0.11439114391143912, + "grad_norm": 5.1460747718811035, + "learning_rate": 9.11764705882353e-06, + "loss": 0.2675, + "step": 93 + }, + { + "epoch": 0.11562115621156212, + "grad_norm": 4.207213878631592, + "learning_rate": 9.215686274509804e-06, + "loss": 0.18, + "step": 94 + }, + { + "epoch": 0.11685116851168512, + "grad_norm": 4.802348613739014, + "learning_rate": 9.31372549019608e-06, + "loss": 0.2891, + "step": 95 + }, + { + "epoch": 0.11808118081180811, + "grad_norm": 4.9332966804504395, + "learning_rate": 9.411764705882354e-06, + "loss": 0.253, + "step": 96 + }, + { + "epoch": 0.11931119311193111, + "grad_norm": 5.841371536254883, + "learning_rate": 9.509803921568628e-06, + "loss": 0.3481, + "step": 97 + }, + { + "epoch": 0.12054120541205413, + "grad_norm": 3.70485782623291, + "learning_rate": 9.607843137254903e-06, + "loss": 0.1688, + "step": 98 + }, + { + "epoch": 0.12177121771217712, + "grad_norm": 4.415471076965332, + "learning_rate": 9.705882352941177e-06, + "loss": 0.1808, + "step": 99 + }, + { + "epoch": 0.12300123001230012, + "grad_norm": 5.058602809906006, + "learning_rate": 9.803921568627451e-06, + "loss": 0.2821, + "step": 100 + }, + { + "epoch": 0.12423124231242312, + "grad_norm": 4.303729057312012, + "learning_rate": 9.901960784313727e-06, + "loss": 0.1856, + "step": 101 + }, + { + "epoch": 0.12546125461254612, + "grad_norm": 4.048065185546875, + "learning_rate": 1e-05, + "loss": 0.1441, + "step": 102 + }, + { + "epoch": 0.12669126691266913, + "grad_norm": 4.463968753814697, + "learning_rate": 1.0098039215686275e-05, + "loss": 0.226, + "step": 103 + }, + { + "epoch": 0.12792127921279212, + "grad_norm": 3.401120901107788, + "learning_rate": 1.0196078431372549e-05, + "loss": 0.1662, + "step": 104 + }, + { + "epoch": 0.12915129151291513, + "grad_norm": 4.119345188140869, + "learning_rate": 1.0294117647058823e-05, + "loss": 0.2043, + "step": 105 + }, + { + "epoch": 0.12915129151291513, + "eval_loss": 0.11874283850193024, + "eval_runtime": 54.7282, + "eval_samples_per_second": 27.518, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8767520821963045, + "eval_sts-test_pearson_dot": 0.8648481444888331, + "eval_sts-test_pearson_euclidean": 0.9053937497921556, + "eval_sts-test_pearson_manhattan": 0.9071737646452815, + "eval_sts-test_pearson_max": 0.9071737646452815, + "eval_sts-test_spearman_cosine": 0.9050983787571032, + "eval_sts-test_spearman_dot": 0.8730474805973213, + "eval_sts-test_spearman_euclidean": 0.9035385735413058, + "eval_sts-test_spearman_manhattan": 0.9054231834122819, + "eval_sts-test_spearman_max": 0.9054231834122819, + "step": 105 + }, + { + "epoch": 0.13038130381303814, + "grad_norm": 5.826413154602051, + "learning_rate": 1.03921568627451e-05, + "loss": 0.3907, + "step": 106 + }, + { + "epoch": 0.13161131611316113, + "grad_norm": 3.2629737854003906, + "learning_rate": 1.0490196078431374e-05, + "loss": 0.1332, + "step": 107 + }, + { + "epoch": 0.13284132841328414, + "grad_norm": 4.044755458831787, + "learning_rate": 1.0588235294117648e-05, + "loss": 0.2243, + "step": 108 + }, + { + "epoch": 0.13407134071340712, + "grad_norm": 3.9784040451049805, + "learning_rate": 1.0686274509803922e-05, + "loss": 0.162, + "step": 109 + }, + { + "epoch": 0.13530135301353013, + "grad_norm": 3.1851444244384766, + "learning_rate": 1.0784313725490196e-05, + "loss": 0.1481, + "step": 110 + }, + { + "epoch": 0.13653136531365315, + "grad_norm": 4.281413555145264, + "learning_rate": 1.0882352941176471e-05, + "loss": 0.2163, + "step": 111 + }, + { + "epoch": 0.13776137761377613, + "grad_norm": 4.62849235534668, + "learning_rate": 1.0980392156862747e-05, + "loss": 0.24, + "step": 112 + }, + { + "epoch": 0.13899138991389914, + "grad_norm": 3.92616868019104, + "learning_rate": 1.1078431372549021e-05, + "loss": 0.1406, + "step": 113 + }, + { + "epoch": 0.14022140221402213, + "grad_norm": 3.8505780696868896, + "learning_rate": 1.1176470588235295e-05, + "loss": 0.1522, + "step": 114 + }, + { + "epoch": 0.14145141451414514, + "grad_norm": 5.220509052276611, + "learning_rate": 1.1274509803921569e-05, + "loss": 0.2593, + "step": 115 + }, + { + "epoch": 0.14268142681426815, + "grad_norm": 4.459743499755859, + "learning_rate": 1.1372549019607844e-05, + "loss": 0.2426, + "step": 116 + }, + { + "epoch": 0.14391143911439114, + "grad_norm": 4.434360504150391, + "learning_rate": 1.1470588235294118e-05, + "loss": 0.1781, + "step": 117 + }, + { + "epoch": 0.14514145141451415, + "grad_norm": 4.638584613800049, + "learning_rate": 1.1568627450980394e-05, + "loss": 0.264, + "step": 118 + }, + { + "epoch": 0.14637146371463713, + "grad_norm": 4.5364484786987305, + "learning_rate": 1.1666666666666668e-05, + "loss": 0.1944, + "step": 119 + }, + { + "epoch": 0.14760147601476015, + "grad_norm": 3.597980499267578, + "learning_rate": 1.1764705882352942e-05, + "loss": 0.1341, + "step": 120 + }, + { + "epoch": 0.14883148831488316, + "grad_norm": 3.5174648761749268, + "learning_rate": 1.1862745098039217e-05, + "loss": 0.155, + "step": 121 + }, + { + "epoch": 0.15006150061500614, + "grad_norm": 4.771029949188232, + "learning_rate": 1.1960784313725491e-05, + "loss": 0.2052, + "step": 122 + }, + { + "epoch": 0.15129151291512916, + "grad_norm": 4.15376615524292, + "learning_rate": 1.2058823529411765e-05, + "loss": 0.2023, + "step": 123 + }, + { + "epoch": 0.15252152521525214, + "grad_norm": 3.5796732902526855, + "learning_rate": 1.215686274509804e-05, + "loss": 0.1519, + "step": 124 + }, + { + "epoch": 0.15375153751537515, + "grad_norm": 3.759777545928955, + "learning_rate": 1.2254901960784315e-05, + "loss": 0.2118, + "step": 125 + }, + { + "epoch": 0.15498154981549817, + "grad_norm": 4.691242218017578, + "learning_rate": 1.235294117647059e-05, + "loss": 0.2489, + "step": 126 + }, + { + "epoch": 0.15498154981549817, + "eval_loss": 0.11467884480953217, + "eval_runtime": 54.6969, + "eval_samples_per_second": 27.534, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8783763084873629, + "eval_sts-test_pearson_dot": 0.8674040012483692, + "eval_sts-test_pearson_euclidean": 0.9069725735634968, + "eval_sts-test_pearson_manhattan": 0.908783443457056, + "eval_sts-test_pearson_max": 0.908783443457056, + "eval_sts-test_spearman_cosine": 0.9058364613112314, + "eval_sts-test_spearman_dot": 0.8738751104254939, + "eval_sts-test_spearman_euclidean": 0.904588080123457, + "eval_sts-test_spearman_manhattan": 0.9067583820471556, + "eval_sts-test_spearman_max": 0.9067583820471556, + "step": 126 + }, + { + "epoch": 0.15621156211562115, + "grad_norm": 4.087611198425293, + "learning_rate": 1.2450980392156864e-05, + "loss": 0.1988, + "step": 127 + }, + { + "epoch": 0.15744157441574416, + "grad_norm": 3.754612684249878, + "learning_rate": 1.2549019607843138e-05, + "loss": 0.1541, + "step": 128 + }, + { + "epoch": 0.15867158671586715, + "grad_norm": 3.9258835315704346, + "learning_rate": 1.2647058823529412e-05, + "loss": 0.1819, + "step": 129 + }, + { + "epoch": 0.15990159901599016, + "grad_norm": 3.88478422164917, + "learning_rate": 1.2745098039215686e-05, + "loss": 0.1582, + "step": 130 + }, + { + "epoch": 0.16113161131611317, + "grad_norm": 4.9845428466796875, + "learning_rate": 1.2843137254901964e-05, + "loss": 0.2866, + "step": 131 + }, + { + "epoch": 0.16236162361623616, + "grad_norm": 4.692960262298584, + "learning_rate": 1.2941176470588238e-05, + "loss": 0.2766, + "step": 132 + }, + { + "epoch": 0.16359163591635917, + "grad_norm": 3.9432125091552734, + "learning_rate": 1.3039215686274511e-05, + "loss": 0.1299, + "step": 133 + }, + { + "epoch": 0.16482164821648215, + "grad_norm": 4.439709663391113, + "learning_rate": 1.3137254901960785e-05, + "loss": 0.2558, + "step": 134 + }, + { + "epoch": 0.16605166051660517, + "grad_norm": 3.631169319152832, + "learning_rate": 1.323529411764706e-05, + "loss": 0.1687, + "step": 135 + }, + { + "epoch": 0.16728167281672818, + "grad_norm": 4.130221843719482, + "learning_rate": 1.3333333333333333e-05, + "loss": 0.173, + "step": 136 + }, + { + "epoch": 0.16851168511685116, + "grad_norm": 4.169937610626221, + "learning_rate": 1.3431372549019607e-05, + "loss": 0.2276, + "step": 137 + }, + { + "epoch": 0.16974169741697417, + "grad_norm": 4.4349751472473145, + "learning_rate": 1.3529411764705885e-05, + "loss": 0.2174, + "step": 138 + }, + { + "epoch": 0.17097170971709716, + "grad_norm": 4.688521862030029, + "learning_rate": 1.3627450980392158e-05, + "loss": 0.2666, + "step": 139 + }, + { + "epoch": 0.17220172201722017, + "grad_norm": 3.7199971675872803, + "learning_rate": 1.3725490196078432e-05, + "loss": 0.1524, + "step": 140 + }, + { + "epoch": 0.17343173431734318, + "grad_norm": 2.8609495162963867, + "learning_rate": 1.3823529411764706e-05, + "loss": 0.1179, + "step": 141 + }, + { + "epoch": 0.17466174661746617, + "grad_norm": 4.374091625213623, + "learning_rate": 1.392156862745098e-05, + "loss": 0.2475, + "step": 142 + }, + { + "epoch": 0.17589175891758918, + "grad_norm": 5.200084209442139, + "learning_rate": 1.4019607843137256e-05, + "loss": 0.2662, + "step": 143 + }, + { + "epoch": 0.17712177121771217, + "grad_norm": 3.720994710922241, + "learning_rate": 1.4117647058823532e-05, + "loss": 0.1596, + "step": 144 + }, + { + "epoch": 0.17835178351783518, + "grad_norm": 3.991046905517578, + "learning_rate": 1.4215686274509805e-05, + "loss": 0.2331, + "step": 145 + }, + { + "epoch": 0.1795817958179582, + "grad_norm": 4.76691198348999, + "learning_rate": 1.431372549019608e-05, + "loss": 0.2905, + "step": 146 + }, + { + "epoch": 0.18081180811808117, + "grad_norm": 3.6453163623809814, + "learning_rate": 1.4411764705882353e-05, + "loss": 0.1342, + "step": 147 + }, + { + "epoch": 0.18081180811808117, + "eval_loss": 0.10875426232814789, + "eval_runtime": 54.7153, + "eval_samples_per_second": 27.524, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8768801472189502, + "eval_sts-test_pearson_dot": 0.8620776961156391, + "eval_sts-test_pearson_euclidean": 0.9073367408863471, + "eval_sts-test_pearson_manhattan": 0.9086519830687241, + "eval_sts-test_pearson_max": 0.9086519830687241, + "eval_sts-test_spearman_cosine": 0.905147068129497, + "eval_sts-test_spearman_dot": 0.869011811845045, + "eval_sts-test_spearman_euclidean": 0.9050077574527855, + "eval_sts-test_spearman_manhattan": 0.9068115017944273, + "eval_sts-test_spearman_max": 0.9068115017944273, + "step": 147 + }, + { + "epoch": 0.1820418204182042, + "grad_norm": 2.6028358936309814, + "learning_rate": 1.4509803921568629e-05, + "loss": 0.0839, + "step": 148 + }, + { + "epoch": 0.18327183271832717, + "grad_norm": 4.445943832397461, + "learning_rate": 1.4607843137254903e-05, + "loss": 0.2055, + "step": 149 + }, + { + "epoch": 0.18450184501845018, + "grad_norm": 4.500098705291748, + "learning_rate": 1.4705882352941179e-05, + "loss": 0.2196, + "step": 150 + }, + { + "epoch": 0.1857318573185732, + "grad_norm": 4.317416667938232, + "learning_rate": 1.4803921568627453e-05, + "loss": 0.2283, + "step": 151 + }, + { + "epoch": 0.18696186961869618, + "grad_norm": 4.395689010620117, + "learning_rate": 1.4901960784313726e-05, + "loss": 0.2105, + "step": 152 + }, + { + "epoch": 0.1881918819188192, + "grad_norm": 3.5757391452789307, + "learning_rate": 1.5000000000000002e-05, + "loss": 0.1534, + "step": 153 + }, + { + "epoch": 0.18942189421894218, + "grad_norm": 3.860861301422119, + "learning_rate": 1.5098039215686276e-05, + "loss": 0.1954, + "step": 154 + }, + { + "epoch": 0.1906519065190652, + "grad_norm": 3.4191622734069824, + "learning_rate": 1.519607843137255e-05, + "loss": 0.1332, + "step": 155 + }, + { + "epoch": 0.1918819188191882, + "grad_norm": 3.8505654335021973, + "learning_rate": 1.5294117647058822e-05, + "loss": 0.19, + "step": 156 + }, + { + "epoch": 0.1931119311193112, + "grad_norm": 4.127209663391113, + "learning_rate": 1.53921568627451e-05, + "loss": 0.1878, + "step": 157 + }, + { + "epoch": 0.1943419434194342, + "grad_norm": 3.7976646423339844, + "learning_rate": 1.5490196078431373e-05, + "loss": 0.1518, + "step": 158 + }, + { + "epoch": 0.19557195571955718, + "grad_norm": 4.613111972808838, + "learning_rate": 1.558823529411765e-05, + "loss": 0.1906, + "step": 159 + }, + { + "epoch": 0.1968019680196802, + "grad_norm": 3.911393880844116, + "learning_rate": 1.568627450980392e-05, + "loss": 0.155, + "step": 160 + }, + { + "epoch": 0.1980319803198032, + "grad_norm": 3.694939374923706, + "learning_rate": 1.5784313725490197e-05, + "loss": 0.1519, + "step": 161 + }, + { + "epoch": 0.1992619926199262, + "grad_norm": 4.334694862365723, + "learning_rate": 1.5882352941176473e-05, + "loss": 0.1726, + "step": 162 + }, + { + "epoch": 0.2004920049200492, + "grad_norm": 3.6630055904388428, + "learning_rate": 1.5980392156862748e-05, + "loss": 0.1618, + "step": 163 + }, + { + "epoch": 0.2017220172201722, + "grad_norm": 4.7789130210876465, + "learning_rate": 1.607843137254902e-05, + "loss": 0.2767, + "step": 164 + }, + { + "epoch": 0.2029520295202952, + "grad_norm": 4.171343803405762, + "learning_rate": 1.6176470588235296e-05, + "loss": 0.1996, + "step": 165 + }, + { + "epoch": 0.20418204182041821, + "grad_norm": 4.386513710021973, + "learning_rate": 1.627450980392157e-05, + "loss": 0.1907, + "step": 166 + }, + { + "epoch": 0.2054120541205412, + "grad_norm": 4.183532238006592, + "learning_rate": 1.6372549019607844e-05, + "loss": 0.1928, + "step": 167 + }, + { + "epoch": 0.2066420664206642, + "grad_norm": 3.8950257301330566, + "learning_rate": 1.647058823529412e-05, + "loss": 0.1507, + "step": 168 + }, + { + "epoch": 0.2066420664206642, + "eval_loss": 0.10821738839149475, + "eval_runtime": 54.7389, + "eval_samples_per_second": 27.512, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8772991124680096, + "eval_sts-test_pearson_dot": 0.861322579093208, + "eval_sts-test_pearson_euclidean": 0.9072621517681675, + "eval_sts-test_pearson_manhattan": 0.9086600802981594, + "eval_sts-test_pearson_max": 0.9086600802981594, + "eval_sts-test_spearman_cosine": 0.9044609865411055, + "eval_sts-test_spearman_dot": 0.8661539962925903, + "eval_sts-test_spearman_euclidean": 0.904084091417667, + "eval_sts-test_spearman_manhattan": 0.9054917423447336, + "eval_sts-test_spearman_max": 0.9054917423447336, + "step": 168 + }, + { + "epoch": 0.2078720787207872, + "grad_norm": 4.002283096313477, + "learning_rate": 1.6568627450980395e-05, + "loss": 0.1637, + "step": 169 + }, + { + "epoch": 0.2091020910209102, + "grad_norm": 4.142872333526611, + "learning_rate": 1.6666666666666667e-05, + "loss": 0.1687, + "step": 170 + }, + { + "epoch": 0.21033210332103322, + "grad_norm": 4.345719337463379, + "learning_rate": 1.6764705882352943e-05, + "loss": 0.2181, + "step": 171 + }, + { + "epoch": 0.2115621156211562, + "grad_norm": 3.7364888191223145, + "learning_rate": 1.686274509803922e-05, + "loss": 0.1496, + "step": 172 + }, + { + "epoch": 0.21279212792127922, + "grad_norm": 4.202157974243164, + "learning_rate": 1.696078431372549e-05, + "loss": 0.1749, + "step": 173 + }, + { + "epoch": 0.2140221402214022, + "grad_norm": 4.639451503753662, + "learning_rate": 1.7058823529411767e-05, + "loss": 0.2374, + "step": 174 + }, + { + "epoch": 0.21525215252152521, + "grad_norm": 4.011781215667725, + "learning_rate": 1.715686274509804e-05, + "loss": 0.2122, + "step": 175 + }, + { + "epoch": 0.21648216482164823, + "grad_norm": 4.113095760345459, + "learning_rate": 1.7254901960784314e-05, + "loss": 0.1617, + "step": 176 + }, + { + "epoch": 0.2177121771217712, + "grad_norm": 4.0442681312561035, + "learning_rate": 1.735294117647059e-05, + "loss": 0.168, + "step": 177 + }, + { + "epoch": 0.21894218942189422, + "grad_norm": 4.375425338745117, + "learning_rate": 1.7450980392156866e-05, + "loss": 0.263, + "step": 178 + }, + { + "epoch": 0.2201722017220172, + "grad_norm": 3.2303390502929688, + "learning_rate": 1.7549019607843138e-05, + "loss": 0.1328, + "step": 179 + }, + { + "epoch": 0.22140221402214022, + "grad_norm": 4.832092761993408, + "learning_rate": 1.7647058823529414e-05, + "loss": 0.3157, + "step": 180 + }, + { + "epoch": 0.22263222632226323, + "grad_norm": 3.57254695892334, + "learning_rate": 1.7745098039215686e-05, + "loss": 0.2164, + "step": 181 + }, + { + "epoch": 0.22386223862238622, + "grad_norm": 3.135535717010498, + "learning_rate": 1.7843137254901965e-05, + "loss": 0.1255, + "step": 182 + }, + { + "epoch": 0.22509225092250923, + "grad_norm": 4.719324588775635, + "learning_rate": 1.7941176470588237e-05, + "loss": 0.2863, + "step": 183 + }, + { + "epoch": 0.22632226322263221, + "grad_norm": 3.8961801528930664, + "learning_rate": 1.8039215686274513e-05, + "loss": 0.155, + "step": 184 + }, + { + "epoch": 0.22755227552275523, + "grad_norm": 2.8389103412628174, + "learning_rate": 1.8137254901960785e-05, + "loss": 0.1271, + "step": 185 + }, + { + "epoch": 0.22878228782287824, + "grad_norm": 4.103536128997803, + "learning_rate": 1.823529411764706e-05, + "loss": 0.216, + "step": 186 + }, + { + "epoch": 0.23001230012300122, + "grad_norm": 4.006705284118652, + "learning_rate": 1.8333333333333333e-05, + "loss": 0.205, + "step": 187 + }, + { + "epoch": 0.23124231242312424, + "grad_norm": 3.424255847930908, + "learning_rate": 1.843137254901961e-05, + "loss": 0.1575, + "step": 188 + }, + { + "epoch": 0.23247232472324722, + "grad_norm": 4.568851947784424, + "learning_rate": 1.8529411764705884e-05, + "loss": 0.1939, + "step": 189 + }, + { + "epoch": 0.23247232472324722, + "eval_loss": 0.1056687980890274, + "eval_runtime": 54.7136, + "eval_samples_per_second": 27.525, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8789160692756717, + "eval_sts-test_pearson_dot": 0.8639029174125306, + "eval_sts-test_pearson_euclidean": 0.9084173029414142, + "eval_sts-test_pearson_manhattan": 0.9093131544369648, + "eval_sts-test_pearson_max": 0.9093131544369648, + "eval_sts-test_spearman_cosine": 0.904571298400435, + "eval_sts-test_spearman_dot": 0.8658778810098052, + "eval_sts-test_spearman_euclidean": 0.9046812074984125, + "eval_sts-test_spearman_manhattan": 0.9056302027474785, + "eval_sts-test_spearman_max": 0.9056302027474785, + "step": 189 + }, + { + "epoch": 0.23370233702337023, + "grad_norm": 3.5909903049468994, + "learning_rate": 1.862745098039216e-05, + "loss": 0.2209, + "step": 190 + }, + { + "epoch": 0.23493234932349324, + "grad_norm": 3.443946361541748, + "learning_rate": 1.8725490196078432e-05, + "loss": 0.153, + "step": 191 + }, + { + "epoch": 0.23616236162361623, + "grad_norm": 3.8604445457458496, + "learning_rate": 1.8823529411764708e-05, + "loss": 0.2187, + "step": 192 + }, + { + "epoch": 0.23739237392373924, + "grad_norm": 3.5916690826416016, + "learning_rate": 1.892156862745098e-05, + "loss": 0.1593, + "step": 193 + }, + { + "epoch": 0.23862238622386223, + "grad_norm": 3.8676974773406982, + "learning_rate": 1.9019607843137255e-05, + "loss": 0.173, + "step": 194 + }, + { + "epoch": 0.23985239852398524, + "grad_norm": 4.338643550872803, + "learning_rate": 1.911764705882353e-05, + "loss": 0.2377, + "step": 195 + }, + { + "epoch": 0.24108241082410825, + "grad_norm": 4.509932994842529, + "learning_rate": 1.9215686274509807e-05, + "loss": 0.2281, + "step": 196 + }, + { + "epoch": 0.24231242312423124, + "grad_norm": 4.282917022705078, + "learning_rate": 1.931372549019608e-05, + "loss": 0.2651, + "step": 197 + }, + { + "epoch": 0.24354243542435425, + "grad_norm": 3.1566977500915527, + "learning_rate": 1.9411764705882355e-05, + "loss": 0.118, + "step": 198 + }, + { + "epoch": 0.24477244772447723, + "grad_norm": 4.118341445922852, + "learning_rate": 1.950980392156863e-05, + "loss": 0.1728, + "step": 199 + }, + { + "epoch": 0.24600246002460024, + "grad_norm": 4.250949859619141, + "learning_rate": 1.9607843137254903e-05, + "loss": 0.2299, + "step": 200 + }, + { + "epoch": 0.24723247232472326, + "grad_norm": 4.084754943847656, + "learning_rate": 1.9705882352941178e-05, + "loss": 0.2342, + "step": 201 + }, + { + "epoch": 0.24846248462484624, + "grad_norm": 3.939434051513672, + "learning_rate": 1.9803921568627454e-05, + "loss": 0.2413, + "step": 202 + }, + { + "epoch": 0.24969249692496925, + "grad_norm": 3.9612276554107666, + "learning_rate": 1.9901960784313726e-05, + "loss": 0.168, + "step": 203 + }, + { + "epoch": 0.25092250922509224, + "grad_norm": 3.401622772216797, + "learning_rate": 2e-05, + "loss": 0.1474, + "step": 204 + }, + { + "epoch": 0.2521525215252153, + "grad_norm": 3.2245850563049316, + "learning_rate": 1.9998802517966852e-05, + "loss": 0.1102, + "step": 205 + }, + { + "epoch": 0.25338253382533826, + "grad_norm": 4.254729270935059, + "learning_rate": 1.9995210358660037e-05, + "loss": 0.2326, + "step": 206 + }, + { + "epoch": 0.25461254612546125, + "grad_norm": 3.603159189224243, + "learning_rate": 1.9989224382388813e-05, + "loss": 0.1787, + "step": 207 + }, + { + "epoch": 0.25584255842558423, + "grad_norm": 3.434582471847534, + "learning_rate": 1.9980846022772978e-05, + "loss": 0.1423, + "step": 208 + }, + { + "epoch": 0.2570725707257073, + "grad_norm": 3.8560950756073, + "learning_rate": 1.997007728639956e-05, + "loss": 0.2069, + "step": 209 + }, + { + "epoch": 0.25830258302583026, + "grad_norm": 3.4417314529418945, + "learning_rate": 1.9956920752342226e-05, + "loss": 0.136, + "step": 210 + }, + { + "epoch": 0.25830258302583026, + "eval_loss": 0.10401736944913864, + "eval_runtime": 54.8034, + "eval_samples_per_second": 27.48, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8792920980078169, + "eval_sts-test_pearson_dot": 0.8612401830255582, + "eval_sts-test_pearson_euclidean": 0.9094380100842928, + "eval_sts-test_pearson_manhattan": 0.9095661408662257, + "eval_sts-test_pearson_max": 0.9095661408662257, + "eval_sts-test_spearman_cosine": 0.905583034917972, + "eval_sts-test_spearman_dot": 0.8658094563311378, + "eval_sts-test_spearman_euclidean": 0.906560626223067, + "eval_sts-test_spearman_manhattan": 0.906644400584392, + "eval_sts-test_spearman_max": 0.906644400584392, + "step": 210 + }, + { + "epoch": 0.25953259532595324, + "grad_norm": 4.170388221740723, + "learning_rate": 1.9941379571543597e-05, + "loss": 0.2407, + "step": 211 + }, + { + "epoch": 0.2607626076260763, + "grad_norm": 4.218827247619629, + "learning_rate": 1.9923457466060637e-05, + "loss": 0.212, + "step": 212 + }, + { + "epoch": 0.26199261992619927, + "grad_norm": 3.6592209339141846, + "learning_rate": 1.9903158728173206e-05, + "loss": 0.1361, + "step": 213 + }, + { + "epoch": 0.26322263222632225, + "grad_norm": 4.208631992340088, + "learning_rate": 1.9880488219356086e-05, + "loss": 0.2356, + "step": 214 + }, + { + "epoch": 0.2644526445264453, + "grad_norm": 2.9232637882232666, + "learning_rate": 1.9855451369114677e-05, + "loss": 0.1059, + "step": 215 + }, + { + "epoch": 0.2656826568265683, + "grad_norm": 4.299160480499268, + "learning_rate": 1.9828054173684646e-05, + "loss": 0.2501, + "step": 216 + }, + { + "epoch": 0.26691266912669126, + "grad_norm": 4.013469219207764, + "learning_rate": 1.9798303194595846e-05, + "loss": 0.1817, + "step": 217 + }, + { + "epoch": 0.26814268142681424, + "grad_norm": 3.691553831100464, + "learning_rate": 1.976620555710087e-05, + "loss": 0.2022, + "step": 218 + }, + { + "epoch": 0.2693726937269373, + "grad_norm": 4.433103561401367, + "learning_rate": 1.973176894846855e-05, + "loss": 0.2235, + "step": 219 + }, + { + "epoch": 0.27060270602706027, + "grad_norm": 4.862768173217773, + "learning_rate": 1.9695001616142916e-05, + "loss": 0.2437, + "step": 220 + }, + { + "epoch": 0.27183271832718325, + "grad_norm": 3.9157614707946777, + "learning_rate": 1.965591236576794e-05, + "loss": 0.1859, + "step": 221 + }, + { + "epoch": 0.2730627306273063, + "grad_norm": 4.705247402191162, + "learning_rate": 1.9614510559078626e-05, + "loss": 0.2167, + "step": 222 + }, + { + "epoch": 0.2742927429274293, + "grad_norm": 3.890500068664551, + "learning_rate": 1.95708061116589e-05, + "loss": 0.1495, + "step": 223 + }, + { + "epoch": 0.27552275522755226, + "grad_norm": 4.393867492675781, + "learning_rate": 1.9524809490566878e-05, + "loss": 0.2876, + "step": 224 + }, + { + "epoch": 0.2767527675276753, + "grad_norm": 3.782416582107544, + "learning_rate": 1.9476531711828027e-05, + "loss": 0.1842, + "step": 225 + }, + { + "epoch": 0.2779827798277983, + "grad_norm": 3.32236647605896, + "learning_rate": 1.942598433779687e-05, + "loss": 0.144, + "step": 226 + }, + { + "epoch": 0.27921279212792127, + "grad_norm": 3.9284870624542236, + "learning_rate": 1.9373179474387858e-05, + "loss": 0.1571, + "step": 227 + }, + { + "epoch": 0.28044280442804426, + "grad_norm": 3.847404956817627, + "learning_rate": 1.9318129768176033e-05, + "loss": 0.209, + "step": 228 + }, + { + "epoch": 0.2816728167281673, + "grad_norm": 4.21238899230957, + "learning_rate": 1.926084840336821e-05, + "loss": 0.2075, + "step": 229 + }, + { + "epoch": 0.2829028290282903, + "grad_norm": 4.167908191680908, + "learning_rate": 1.9201349098645433e-05, + "loss": 0.1722, + "step": 230 + }, + { + "epoch": 0.28413284132841327, + "grad_norm": 3.7701351642608643, + "learning_rate": 1.9139646103877378e-05, + "loss": 0.1464, + "step": 231 + }, + { + "epoch": 0.28413284132841327, + "eval_loss": 0.10392418503761292, + "eval_runtime": 54.7341, + "eval_samples_per_second": 27.515, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8822954324429473, + "eval_sts-test_pearson_dot": 0.8642863367586305, + "eval_sts-test_pearson_euclidean": 0.9122889538029727, + "eval_sts-test_pearson_manhattan": 0.912099304650421, + "eval_sts-test_pearson_max": 0.9122889538029727, + "eval_sts-test_spearman_cosine": 0.9087387596175093, + "eval_sts-test_spearman_dot": 0.8704923178256567, + "eval_sts-test_spearman_euclidean": 0.9097842833373965, + "eval_sts-test_spearman_manhattan": 0.9095372563745162, + "eval_sts-test_spearman_max": 0.9097842833373965, + "step": 231 + }, + { + "epoch": 0.2853628536285363, + "grad_norm": 4.548703193664551, + "learning_rate": 1.9075754196709574e-05, + "loss": 0.2675, + "step": 232 + }, + { + "epoch": 0.2865928659286593, + "grad_norm": 5.041469573974609, + "learning_rate": 1.900968867902419e-05, + "loss": 0.2585, + "step": 233 + }, + { + "epoch": 0.2878228782287823, + "grad_norm": 3.0036237239837646, + "learning_rate": 1.894146537327533e-05, + "loss": 0.134, + "step": 234 + }, + { + "epoch": 0.2890528905289053, + "grad_norm": 3.6082603931427, + "learning_rate": 1.8871100618699553e-05, + "loss": 0.1765, + "step": 235 + }, + { + "epoch": 0.2902829028290283, + "grad_norm": 3.8336241245269775, + "learning_rate": 1.8798611267402745e-05, + "loss": 0.1826, + "step": 236 + }, + { + "epoch": 0.2915129151291513, + "grad_norm": 4.307932376861572, + "learning_rate": 1.872401468032406e-05, + "loss": 0.222, + "step": 237 + }, + { + "epoch": 0.29274292742927427, + "grad_norm": 3.153963088989258, + "learning_rate": 1.864732872307804e-05, + "loss": 0.134, + "step": 238 + }, + { + "epoch": 0.2939729397293973, + "grad_norm": 4.044833660125732, + "learning_rate": 1.8568571761675893e-05, + "loss": 0.1902, + "step": 239 + }, + { + "epoch": 0.2952029520295203, + "grad_norm": 4.640310287475586, + "learning_rate": 1.8487762658126872e-05, + "loss": 0.2461, + "step": 240 + }, + { + "epoch": 0.2964329643296433, + "grad_norm": 4.932340145111084, + "learning_rate": 1.8404920765920898e-05, + "loss": 0.3094, + "step": 241 + }, + { + "epoch": 0.2976629766297663, + "grad_norm": 4.0233917236328125, + "learning_rate": 1.8320065925393468e-05, + "loss": 0.2252, + "step": 242 + }, + { + "epoch": 0.2988929889298893, + "grad_norm": 4.369536399841309, + "learning_rate": 1.8233218458973984e-05, + "loss": 0.2466, + "step": 243 + }, + { + "epoch": 0.3001230012300123, + "grad_norm": 3.6295106410980225, + "learning_rate": 1.814439916631857e-05, + "loss": 0.139, + "step": 244 + }, + { + "epoch": 0.3013530135301353, + "grad_norm": 3.705105781555176, + "learning_rate": 1.8053629319328662e-05, + "loss": 0.154, + "step": 245 + }, + { + "epoch": 0.3025830258302583, + "grad_norm": 3.7480130195617676, + "learning_rate": 1.796093065705644e-05, + "loss": 0.1979, + "step": 246 + }, + { + "epoch": 0.3038130381303813, + "grad_norm": 2.5885541439056396, + "learning_rate": 1.786632538049842e-05, + "loss": 0.1121, + "step": 247 + }, + { + "epoch": 0.3050430504305043, + "grad_norm": 3.3691048622131348, + "learning_rate": 1.7769836147278385e-05, + "loss": 0.1361, + "step": 248 + }, + { + "epoch": 0.3062730627306273, + "grad_norm": 4.20883321762085, + "learning_rate": 1.7671486066220965e-05, + "loss": 0.2492, + "step": 249 + }, + { + "epoch": 0.3075030750307503, + "grad_norm": 3.8119523525238037, + "learning_rate": 1.757129869181718e-05, + "loss": 0.1903, + "step": 250 + }, + { + "epoch": 0.3087330873308733, + "grad_norm": 4.464923858642578, + "learning_rate": 1.746929801858317e-05, + "loss": 0.2333, + "step": 251 + }, + { + "epoch": 0.30996309963099633, + "grad_norm": 4.029540061950684, + "learning_rate": 1.736550847531366e-05, + "loss": 0.1805, + "step": 252 + }, + { + "epoch": 0.30996309963099633, + "eval_loss": 0.10298814624547958, + "eval_runtime": 54.69, + "eval_samples_per_second": 27.537, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.881635556166377, + "eval_sts-test_pearson_dot": 0.862389303076424, + "eval_sts-test_pearson_euclidean": 0.9125260227425505, + "eval_sts-test_pearson_manhattan": 0.9128421094636647, + "eval_sts-test_pearson_max": 0.9128421094636647, + "eval_sts-test_spearman_cosine": 0.9098964747497047, + "eval_sts-test_spearman_dot": 0.8698043119330254, + "eval_sts-test_spearman_euclidean": 0.9103645729438589, + "eval_sts-test_spearman_manhattan": 0.9110424650514156, + "eval_sts-test_spearman_max": 0.9110424650514156, + "step": 252 + }, + { + "epoch": 0.3111931119311193, + "grad_norm": 3.990410089492798, + "learning_rate": 1.725995491923131e-05, + "loss": 0.1929, + "step": 253 + }, + { + "epoch": 0.3124231242312423, + "grad_norm": 3.7284581661224365, + "learning_rate": 1.7152662630033506e-05, + "loss": 0.1424, + "step": 254 + }, + { + "epoch": 0.31365313653136534, + "grad_norm": 3.8791370391845703, + "learning_rate": 1.7043657303837965e-05, + "loss": 0.2318, + "step": 255 + }, + { + "epoch": 0.3148831488314883, + "grad_norm": 3.4804205894470215, + "learning_rate": 1.693296504702862e-05, + "loss": 0.1524, + "step": 256 + }, + { + "epoch": 0.3161131611316113, + "grad_norm": 3.573451519012451, + "learning_rate": 1.682061237000322e-05, + "loss": 0.2195, + "step": 257 + }, + { + "epoch": 0.3173431734317343, + "grad_norm": 3.5766184329986572, + "learning_rate": 1.6706626180824185e-05, + "loss": 0.1338, + "step": 258 + }, + { + "epoch": 0.31857318573185733, + "grad_norm": 4.488210201263428, + "learning_rate": 1.659103377877423e-05, + "loss": 0.2543, + "step": 259 + }, + { + "epoch": 0.3198031980319803, + "grad_norm": 4.0144147872924805, + "learning_rate": 1.647386284781828e-05, + "loss": 0.202, + "step": 260 + }, + { + "epoch": 0.3210332103321033, + "grad_norm": 3.4031426906585693, + "learning_rate": 1.6355141449973254e-05, + "loss": 0.1489, + "step": 261 + }, + { + "epoch": 0.32226322263222634, + "grad_norm": 3.8359596729278564, + "learning_rate": 1.6234898018587336e-05, + "loss": 0.1937, + "step": 262 + }, + { + "epoch": 0.3234932349323493, + "grad_norm": 4.457846641540527, + "learning_rate": 1.6113161351530257e-05, + "loss": 0.2334, + "step": 263 + }, + { + "epoch": 0.3247232472324723, + "grad_norm": 4.167722702026367, + "learning_rate": 1.598996060429634e-05, + "loss": 0.1942, + "step": 264 + }, + { + "epoch": 0.32595325953259535, + "grad_norm": 4.352579116821289, + "learning_rate": 1.586532528302183e-05, + "loss": 0.2013, + "step": 265 + }, + { + "epoch": 0.32718327183271834, + "grad_norm": 5.293665409088135, + "learning_rate": 1.5739285237418323e-05, + "loss": 0.2954, + "step": 266 + }, + { + "epoch": 0.3284132841328413, + "grad_norm": 3.7269585132598877, + "learning_rate": 1.5611870653623826e-05, + "loss": 0.188, + "step": 267 + }, + { + "epoch": 0.3296432964329643, + "grad_norm": 3.8485231399536133, + "learning_rate": 1.548311204697331e-05, + "loss": 0.1688, + "step": 268 + }, + { + "epoch": 0.33087330873308735, + "grad_norm": 3.183656692504883, + "learning_rate": 1.5353040254690396e-05, + "loss": 0.1415, + "step": 269 + }, + { + "epoch": 0.33210332103321033, + "grad_norm": 4.292448997497559, + "learning_rate": 1.5221686428501929e-05, + "loss": 0.2249, + "step": 270 + }, + { + "epoch": 0.3333333333333333, + "grad_norm": 4.776716232299805, + "learning_rate": 1.508908202717729e-05, + "loss": 0.2606, + "step": 271 + }, + { + "epoch": 0.33456334563345635, + "grad_norm": 4.753313064575195, + "learning_rate": 1.4955258808994093e-05, + "loss": 0.2559, + "step": 272 + }, + { + "epoch": 0.33579335793357934, + "grad_norm": 4.271998882293701, + "learning_rate": 1.482024882413222e-05, + "loss": 0.2673, + "step": 273 + }, + { + "epoch": 0.33579335793357934, + "eval_loss": 0.10389706492424011, + "eval_runtime": 54.7165, + "eval_samples_per_second": 27.524, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8802537579023317, + "eval_sts-test_pearson_dot": 0.8611279643446735, + "eval_sts-test_pearson_euclidean": 0.9104828974356078, + "eval_sts-test_pearson_manhattan": 0.9112003024801107, + "eval_sts-test_pearson_max": 0.9112003024801107, + "eval_sts-test_spearman_cosine": 0.9077587069930375, + "eval_sts-test_spearman_dot": 0.8683557106354935, + "eval_sts-test_spearman_euclidean": 0.908703271867226, + "eval_sts-test_spearman_manhattan": 0.9088831719380195, + "eval_sts-test_spearman_max": 0.9088831719380195, + "step": 273 + }, + { + "epoch": 0.3370233702337023, + "grad_norm": 3.5790419578552246, + "learning_rate": 1.4684084406997903e-05, + "loss": 0.1618, + "step": 274 + }, + { + "epoch": 0.33825338253382536, + "grad_norm": 4.819033145904541, + "learning_rate": 1.4546798168479756e-05, + "loss": 0.2602, + "step": 275 + }, + { + "epoch": 0.33948339483394835, + "grad_norm": 4.567826747894287, + "learning_rate": 1.4408422988138585e-05, + "loss": 0.2339, + "step": 276 + }, + { + "epoch": 0.34071340713407133, + "grad_norm": 4.182609558105469, + "learning_rate": 1.4268992006332847e-05, + "loss": 0.1843, + "step": 277 + }, + { + "epoch": 0.3419434194341943, + "grad_norm": 3.4522156715393066, + "learning_rate": 1.412853861628166e-05, + "loss": 0.133, + "step": 278 + }, + { + "epoch": 0.34317343173431736, + "grad_norm": 4.6532301902771, + "learning_rate": 1.3987096456067236e-05, + "loss": 0.2345, + "step": 279 + }, + { + "epoch": 0.34440344403444034, + "grad_norm": 4.240933418273926, + "learning_rate": 1.3844699400578696e-05, + "loss": 0.2808, + "step": 280 + }, + { + "epoch": 0.3456334563345633, + "grad_norm": 3.000117063522339, + "learning_rate": 1.3701381553399144e-05, + "loss": 0.1044, + "step": 281 + }, + { + "epoch": 0.34686346863468637, + "grad_norm": 3.7988216876983643, + "learning_rate": 1.3557177238637987e-05, + "loss": 0.1622, + "step": 282 + }, + { + "epoch": 0.34809348093480935, + "grad_norm": 3.2597107887268066, + "learning_rate": 1.3412120992710425e-05, + "loss": 0.1303, + "step": 283 + }, + { + "epoch": 0.34932349323493234, + "grad_norm": 3.2426445484161377, + "learning_rate": 1.3266247556066122e-05, + "loss": 0.1453, + "step": 284 + }, + { + "epoch": 0.3505535055350554, + "grad_norm": 4.482458114624023, + "learning_rate": 1.3119591864868979e-05, + "loss": 0.237, + "step": 285 + }, + { + "epoch": 0.35178351783517836, + "grad_norm": 4.062747478485107, + "learning_rate": 1.2972189042630044e-05, + "loss": 0.1726, + "step": 286 + }, + { + "epoch": 0.35301353013530135, + "grad_norm": 3.9885880947113037, + "learning_rate": 1.2824074391795571e-05, + "loss": 0.2195, + "step": 287 + }, + { + "epoch": 0.35424354243542433, + "grad_norm": 5.205960750579834, + "learning_rate": 1.2675283385292212e-05, + "loss": 0.3016, + "step": 288 + }, + { + "epoch": 0.35547355473554737, + "grad_norm": 3.2820823192596436, + "learning_rate": 1.252585165803135e-05, + "loss": 0.1626, + "step": 289 + }, + { + "epoch": 0.35670356703567035, + "grad_norm": 4.133265495300293, + "learning_rate": 1.2375814998374714e-05, + "loss": 0.1902, + "step": 290 + }, + { + "epoch": 0.35793357933579334, + "grad_norm": 3.349637746810913, + "learning_rate": 1.2225209339563144e-05, + "loss": 0.1387, + "step": 291 + }, + { + "epoch": 0.3591635916359164, + "grad_norm": 2.7458724975585938, + "learning_rate": 1.2074070751110753e-05, + "loss": 0.1047, + "step": 292 + }, + { + "epoch": 0.36039360393603936, + "grad_norm": 3.7697091102600098, + "learning_rate": 1.1922435430166372e-05, + "loss": 0.1954, + "step": 293 + }, + { + "epoch": 0.36162361623616235, + "grad_norm": 4.1529622077941895, + "learning_rate": 1.1770339692844484e-05, + "loss": 0.2089, + "step": 294 + }, + { + "epoch": 0.36162361623616235, + "eval_loss": 0.10294178128242493, + "eval_runtime": 54.5996, + "eval_samples_per_second": 27.583, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8803510241742277, + "eval_sts-test_pearson_dot": 0.8614562911474479, + "eval_sts-test_pearson_euclidean": 0.9105274765595701, + "eval_sts-test_pearson_manhattan": 0.9112776803683604, + "eval_sts-test_pearson_max": 0.9112776803683604, + "eval_sts-test_spearman_cosine": 0.9082726304788564, + "eval_sts-test_spearman_dot": 0.8687116203836315, + "eval_sts-test_spearman_euclidean": 0.9088194013905367, + "eval_sts-test_spearman_manhattan": 0.9091452800759889, + "eval_sts-test_spearman_max": 0.9091452800759889, + "step": 294 + }, + { + "epoch": 0.3628536285362854, + "grad_norm": 3.3324942588806152, + "learning_rate": 1.161781996552765e-05, + "loss": 0.1485, + "step": 295 + }, + { + "epoch": 0.3640836408364084, + "grad_norm": 3.477254867553711, + "learning_rate": 1.1464912776142494e-05, + "loss": 0.1724, + "step": 296 + }, + { + "epoch": 0.36531365313653136, + "grad_norm": 3.933436393737793, + "learning_rate": 1.1311654745411424e-05, + "loss": 0.2017, + "step": 297 + }, + { + "epoch": 0.36654366543665434, + "grad_norm": 3.6212170124053955, + "learning_rate": 1.115808257808209e-05, + "loss": 0.1591, + "step": 298 + }, + { + "epoch": 0.3677736777367774, + "grad_norm": 4.0765700340271, + "learning_rate": 1.1004233054136726e-05, + "loss": 0.2396, + "step": 299 + }, + { + "epoch": 0.36900369003690037, + "grad_norm": 3.589646816253662, + "learning_rate": 1.0850143019983475e-05, + "loss": 0.1395, + "step": 300 + }, + { + "epoch": 0.37023370233702335, + "grad_norm": 3.7769243717193604, + "learning_rate": 1.0695849379631816e-05, + "loss": 0.1806, + "step": 301 + }, + { + "epoch": 0.3714637146371464, + "grad_norm": 3.4720847606658936, + "learning_rate": 1.0541389085854177e-05, + "loss": 0.1882, + "step": 302 + }, + { + "epoch": 0.3726937269372694, + "grad_norm": 2.9006810188293457, + "learning_rate": 1.038679913133589e-05, + "loss": 0.1188, + "step": 303 + }, + { + "epoch": 0.37392373923739236, + "grad_norm": 3.7660434246063232, + "learning_rate": 1.023211653981556e-05, + "loss": 0.1564, + "step": 304 + }, + { + "epoch": 0.3751537515375154, + "grad_norm": 5.082170486450195, + "learning_rate": 1.0077378357218023e-05, + "loss": 0.313, + "step": 305 + }, + { + "epoch": 0.3763837638376384, + "grad_norm": 3.5429434776306152, + "learning_rate": 9.922621642781982e-06, + "loss": 0.1455, + "step": 306 + }, + { + "epoch": 0.37761377613776137, + "grad_norm": 3.1348257064819336, + "learning_rate": 9.767883460184447e-06, + "loss": 0.1535, + "step": 307 + }, + { + "epoch": 0.37884378843788435, + "grad_norm": 2.881880521774292, + "learning_rate": 9.613200868664112e-06, + "loss": 0.099, + "step": 308 + }, + { + "epoch": 0.3800738007380074, + "grad_norm": 3.5104594230651855, + "learning_rate": 9.458610914145826e-06, + "loss": 0.1733, + "step": 309 + }, + { + "epoch": 0.3813038130381304, + "grad_norm": 3.9202194213867188, + "learning_rate": 9.304150620368189e-06, + "loss": 0.1891, + "step": 310 + }, + { + "epoch": 0.38253382533825336, + "grad_norm": 3.655240297317505, + "learning_rate": 9.149856980016529e-06, + "loss": 0.2128, + "step": 311 + }, + { + "epoch": 0.3837638376383764, + "grad_norm": 3.766303062438965, + "learning_rate": 8.995766945863278e-06, + "loss": 0.2042, + "step": 312 + }, + { + "epoch": 0.3849938499384994, + "grad_norm": 3.8122780323028564, + "learning_rate": 8.841917421917913e-06, + "loss": 0.203, + "step": 313 + }, + { + "epoch": 0.3862238622386224, + "grad_norm": 4.085626602172852, + "learning_rate": 8.688345254588579e-06, + "loss": 0.2249, + "step": 314 + }, + { + "epoch": 0.3874538745387454, + "grad_norm": 3.744234323501587, + "learning_rate": 8.53508722385751e-06, + "loss": 0.1597, + "step": 315 + }, + { + "epoch": 0.3874538745387454, + "eval_loss": 0.10140395164489746, + "eval_runtime": 54.6293, + "eval_samples_per_second": 27.568, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8788980244871143, + "eval_sts-test_pearson_dot": 0.8616979928109831, + "eval_sts-test_pearson_euclidean": 0.9090675882181183, + "eval_sts-test_pearson_manhattan": 0.909351159725011, + "eval_sts-test_pearson_max": 0.909351159725011, + "eval_sts-test_spearman_cosine": 0.9074493862743003, + "eval_sts-test_spearman_dot": 0.8701774479505067, + "eval_sts-test_spearman_euclidean": 0.9075559837789346, + "eval_sts-test_spearman_manhattan": 0.9076191725600193, + "eval_sts-test_spearman_max": 0.9076191725600193, + "step": 315 + }, + { + "epoch": 0.3886838868388684, + "grad_norm": 3.4346373081207275, + "learning_rate": 8.382180034472353e-06, + "loss": 0.1358, + "step": 316 + }, + { + "epoch": 0.3899138991389914, + "grad_norm": 3.872002124786377, + "learning_rate": 8.229660307155518e-06, + "loss": 0.207, + "step": 317 + }, + { + "epoch": 0.39114391143911437, + "grad_norm": 3.4921915531158447, + "learning_rate": 8.077564569833633e-06, + "loss": 0.193, + "step": 318 + }, + { + "epoch": 0.3923739237392374, + "grad_norm": 3.2774128913879395, + "learning_rate": 7.92592924888925e-06, + "loss": 0.1141, + "step": 319 + }, + { + "epoch": 0.3936039360393604, + "grad_norm": 4.900540351867676, + "learning_rate": 7.774790660436857e-06, + "loss": 0.2835, + "step": 320 + }, + { + "epoch": 0.3948339483394834, + "grad_norm": 4.073228359222412, + "learning_rate": 7.6241850016252915e-06, + "loss": 0.2589, + "step": 321 + }, + { + "epoch": 0.3960639606396064, + "grad_norm": 2.549339532852173, + "learning_rate": 7.4741483419686475e-06, + "loss": 0.088, + "step": 322 + }, + { + "epoch": 0.3972939729397294, + "grad_norm": 4.411525726318359, + "learning_rate": 7.324716614707792e-06, + "loss": 0.1675, + "step": 323 + }, + { + "epoch": 0.3985239852398524, + "grad_norm": 3.336052656173706, + "learning_rate": 7.175925608204428e-06, + "loss": 0.1525, + "step": 324 + }, + { + "epoch": 0.3997539975399754, + "grad_norm": 3.2689311504364014, + "learning_rate": 7.0278109573699574e-06, + "loss": 0.1401, + "step": 325 + }, + { + "epoch": 0.4009840098400984, + "grad_norm": 3.8623855113983154, + "learning_rate": 6.880408135131022e-06, + "loss": 0.2109, + "step": 326 + }, + { + "epoch": 0.4022140221402214, + "grad_norm": 3.1453464031219482, + "learning_rate": 6.733752443933879e-06, + "loss": 0.1382, + "step": 327 + }, + { + "epoch": 0.4034440344403444, + "grad_norm": 3.09243106842041, + "learning_rate": 6.587879007289576e-06, + "loss": 0.1724, + "step": 328 + }, + { + "epoch": 0.4046740467404674, + "grad_norm": 4.039675712585449, + "learning_rate": 6.442822761362015e-06, + "loss": 0.1668, + "step": 329 + }, + { + "epoch": 0.4059040590405904, + "grad_norm": 3.2855875492095947, + "learning_rate": 6.298618446600856e-06, + "loss": 0.1606, + "step": 330 + }, + { + "epoch": 0.4071340713407134, + "grad_norm": 3.927316665649414, + "learning_rate": 6.155300599421305e-06, + "loss": 0.2102, + "step": 331 + }, + { + "epoch": 0.40836408364083643, + "grad_norm": 3.7205469608306885, + "learning_rate": 6.0129035439327665e-06, + "loss": 0.1737, + "step": 332 + }, + { + "epoch": 0.4095940959409594, + "grad_norm": 3.6478657722473145, + "learning_rate": 5.871461383718344e-06, + "loss": 0.1641, + "step": 333 + }, + { + "epoch": 0.4108241082410824, + "grad_norm": 3.9690299034118652, + "learning_rate": 5.731007993667155e-06, + "loss": 0.1984, + "step": 334 + }, + { + "epoch": 0.41205412054120544, + "grad_norm": 3.357426166534424, + "learning_rate": 5.5915770118614195e-06, + "loss": 0.1395, + "step": 335 + }, + { + "epoch": 0.4132841328413284, + "grad_norm": 3.187108278274536, + "learning_rate": 5.453201831520245e-06, + "loss": 0.1236, + "step": 336 + }, + { + "epoch": 0.4132841328413284, + "eval_loss": 0.10084880888462067, + "eval_runtime": 54.9326, + "eval_samples_per_second": 27.415, + "eval_steps_per_second": 0.218, + "eval_sts-test_pearson_cosine": 0.8786030380146035, + "eval_sts-test_pearson_dot": 0.8615207822521098, + "eval_sts-test_pearson_euclidean": 0.9085208579537902, + "eval_sts-test_pearson_manhattan": 0.908989834558373, + "eval_sts-test_pearson_max": 0.908989834558373, + "eval_sts-test_spearman_cosine": 0.9066104343769958, + "eval_sts-test_spearman_dot": 0.8696513521215895, + "eval_sts-test_spearman_euclidean": 0.906370120103324, + "eval_sts-test_spearman_manhattan": 0.906956674886381, + "eval_sts-test_spearman_max": 0.906956674886381, + "step": 336 + }, + { + "epoch": 0.4145141451414514, + "grad_norm": 3.129255771636963, + "learning_rate": 5.3159155930021e-06, + "loss": 0.1405, + "step": 337 + }, + { + "epoch": 0.4157441574415744, + "grad_norm": 3.2555580139160156, + "learning_rate": 5.17975117586778e-06, + "loss": 0.1461, + "step": 338 + }, + { + "epoch": 0.41697416974169743, + "grad_norm": 3.160216808319092, + "learning_rate": 5.044741191005908e-06, + "loss": 0.1151, + "step": 339 + }, + { + "epoch": 0.4182041820418204, + "grad_norm": 3.260749340057373, + "learning_rate": 4.91091797282271e-06, + "loss": 0.1282, + "step": 340 + }, + { + "epoch": 0.4194341943419434, + "grad_norm": 4.1700286865234375, + "learning_rate": 4.778313571498075e-06, + "loss": 0.2155, + "step": 341 + }, + { + "epoch": 0.42066420664206644, + "grad_norm": 3.473802328109741, + "learning_rate": 4.646959745309606e-06, + "loss": 0.1344, + "step": 342 + }, + { + "epoch": 0.4218942189421894, + "grad_norm": 3.5372884273529053, + "learning_rate": 4.516887953026694e-06, + "loss": 0.1854, + "step": 343 + }, + { + "epoch": 0.4231242312423124, + "grad_norm": 3.766799211502075, + "learning_rate": 4.388129346376177e-06, + "loss": 0.1766, + "step": 344 + }, + { + "epoch": 0.42435424354243545, + "grad_norm": 3.4864675998687744, + "learning_rate": 4.260714762581682e-06, + "loss": 0.122, + "step": 345 + }, + { + "epoch": 0.42558425584255843, + "grad_norm": 3.3038723468780518, + "learning_rate": 4.1346747169781695e-06, + "loss": 0.142, + "step": 346 + }, + { + "epoch": 0.4268142681426814, + "grad_norm": 3.1204986572265625, + "learning_rate": 4.010039395703665e-06, + "loss": 0.1434, + "step": 347 + }, + { + "epoch": 0.4280442804428044, + "grad_norm": 3.8905293941497803, + "learning_rate": 3.886838648469742e-06, + "loss": 0.1687, + "step": 348 + }, + { + "epoch": 0.42927429274292744, + "grad_norm": 3.4462125301361084, + "learning_rate": 3.765101981412669e-06, + "loss": 0.1751, + "step": 349 + }, + { + "epoch": 0.43050430504305043, + "grad_norm": 3.0247697830200195, + "learning_rate": 3.6448585500267487e-06, + "loss": 0.1253, + "step": 350 + }, + { + "epoch": 0.4317343173431734, + "grad_norm": 3.107347249984741, + "learning_rate": 3.5261371521817247e-06, + "loss": 0.1387, + "step": 351 + }, + { + "epoch": 0.43296432964329645, + "grad_norm": 4.09791898727417, + "learning_rate": 3.4089662212257733e-06, + "loss": 0.181, + "step": 352 + }, + { + "epoch": 0.43419434194341944, + "grad_norm": 2.587646722793579, + "learning_rate": 3.2933738191758158e-06, + "loss": 0.101, + "step": 353 + }, + { + "epoch": 0.4354243542435424, + "grad_norm": 3.613358736038208, + "learning_rate": 3.179387629996782e-06, + "loss": 0.1552, + "step": 354 + }, + { + "epoch": 0.43665436654366546, + "grad_norm": 4.2074809074401855, + "learning_rate": 3.067034952971382e-06, + "loss": 0.2676, + "step": 355 + }, + { + "epoch": 0.43788437884378845, + "grad_norm": 3.7506251335144043, + "learning_rate": 2.956342696162037e-06, + "loss": 0.1638, + "step": 356 + }, + { + "epoch": 0.43911439114391143, + "grad_norm": 3.2212603092193604, + "learning_rate": 2.8473373699664965e-06, + "loss": 0.19, + "step": 357 + }, + { + "epoch": 0.43911439114391143, + "eval_loss": 0.10077783465385437, + "eval_runtime": 54.6203, + "eval_samples_per_second": 27.572, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8796435169424439, + "eval_sts-test_pearson_dot": 0.8633927890650783, + "eval_sts-test_pearson_euclidean": 0.9090664236075847, + "eval_sts-test_pearson_manhattan": 0.9095614639035563, + "eval_sts-test_pearson_max": 0.9095614639035563, + "eval_sts-test_spearman_cosine": 0.907205357645933, + "eval_sts-test_spearman_dot": 0.8702227362270348, + "eval_sts-test_spearman_euclidean": 0.9068201387878759, + "eval_sts-test_spearman_manhattan": 0.9070422392981614, + "eval_sts-test_spearman_max": 0.907205357645933, + "step": 357 + }, + { + "epoch": 0.4403444034440344, + "grad_norm": 2.998774528503418, + "learning_rate": 2.740045080768692e-06, + "loss": 0.1152, + "step": 358 + }, + { + "epoch": 0.44157441574415746, + "grad_norm": 3.922896385192871, + "learning_rate": 2.6344915246863378e-06, + "loss": 0.1639, + "step": 359 + }, + { + "epoch": 0.44280442804428044, + "grad_norm": 3.852358818054199, + "learning_rate": 2.530701981416831e-06, + "loss": 0.1624, + "step": 360 + }, + { + "epoch": 0.4440344403444034, + "grad_norm": 4.164193153381348, + "learning_rate": 2.428701308182826e-06, + "loss": 0.203, + "step": 361 + }, + { + "epoch": 0.44526445264452646, + "grad_norm": 4.117493629455566, + "learning_rate": 2.3285139337790343e-06, + "loss": 0.1856, + "step": 362 + }, + { + "epoch": 0.44649446494464945, + "grad_norm": 3.715933322906494, + "learning_rate": 2.2301638527216196e-06, + "loss": 0.1978, + "step": 363 + }, + { + "epoch": 0.44772447724477243, + "grad_norm": 3.033038377761841, + "learning_rate": 2.1336746195015845e-06, + "loss": 0.1457, + "step": 364 + }, + { + "epoch": 0.4489544895448955, + "grad_norm": 3.720082998275757, + "learning_rate": 2.0390693429435626e-06, + "loss": 0.176, + "step": 365 + }, + { + "epoch": 0.45018450184501846, + "grad_norm": 3.170830726623535, + "learning_rate": 1.946370680671341e-06, + "loss": 0.1742, + "step": 366 + }, + { + "epoch": 0.45141451414514144, + "grad_norm": 3.5216987133026123, + "learning_rate": 1.8556008336814302e-06, + "loss": 0.1599, + "step": 367 + }, + { + "epoch": 0.45264452644526443, + "grad_norm": 3.75478196144104, + "learning_rate": 1.7667815410260181e-06, + "loss": 0.2085, + "step": 368 + }, + { + "epoch": 0.45387453874538747, + "grad_norm": 4.367212295532227, + "learning_rate": 1.679934074606533e-06, + "loss": 0.2255, + "step": 369 + }, + { + "epoch": 0.45510455104551045, + "grad_norm": 3.956402063369751, + "learning_rate": 1.5950792340791044e-06, + "loss": 0.1941, + "step": 370 + }, + { + "epoch": 0.45633456334563344, + "grad_norm": 2.1625616550445557, + "learning_rate": 1.5122373418731306e-06, + "loss": 0.0769, + "step": 371 + }, + { + "epoch": 0.4575645756457565, + "grad_norm": 3.6466803550720215, + "learning_rate": 1.4314282383241097e-06, + "loss": 0.2031, + "step": 372 + }, + { + "epoch": 0.45879458794587946, + "grad_norm": 4.083393096923828, + "learning_rate": 1.3526712769219619e-06, + "loss": 0.2151, + "step": 373 + }, + { + "epoch": 0.46002460024600245, + "grad_norm": 3.6854231357574463, + "learning_rate": 1.2759853196759454e-06, + "loss": 0.2115, + "step": 374 + }, + { + "epoch": 0.4612546125461255, + "grad_norm": 3.0205180644989014, + "learning_rate": 1.201388732597255e-06, + "loss": 0.1241, + "step": 375 + }, + { + "epoch": 0.46248462484624847, + "grad_norm": 3.7446978092193604, + "learning_rate": 1.1288993813004467e-06, + "loss": 0.1693, + "step": 376 + }, + { + "epoch": 0.46371463714637146, + "grad_norm": 4.181955337524414, + "learning_rate": 1.0585346267246743e-06, + "loss": 0.2086, + "step": 377 + }, + { + "epoch": 0.46494464944649444, + "grad_norm": 4.160934925079346, + "learning_rate": 9.903113209758098e-07, + "loss": 0.1661, + "step": 378 + }, + { + "epoch": 0.46494464944649444, + "eval_loss": 0.10039300471544266, + "eval_runtime": 54.6221, + "eval_samples_per_second": 27.571, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8801762243717315, + "eval_sts-test_pearson_dot": 0.8639479260425424, + "eval_sts-test_pearson_euclidean": 0.9094698598636811, + "eval_sts-test_pearson_manhattan": 0.9099527175289687, + "eval_sts-test_pearson_max": 0.9099527175289687, + "eval_sts-test_spearman_cosine": 0.9074492520205162, + "eval_sts-test_spearman_dot": 0.8704936603634984, + "eval_sts-test_spearman_euclidean": 0.9069152352183324, + "eval_sts-test_spearman_manhattan": 0.9072497061459718, + "eval_sts-test_spearman_max": 0.9074492520205162, + "step": 378 + }, + { + "epoch": 0.4661746617466175, + "grad_norm": 3.3712666034698486, + "learning_rate": 9.242458032904311e-07, + "loss": 0.1508, + "step": 379 + }, + { + "epoch": 0.46740467404674046, + "grad_norm": 4.186792850494385, + "learning_rate": 8.603538961226243e-07, + "loss": 0.1802, + "step": 380 + }, + { + "epoch": 0.46863468634686345, + "grad_norm": 2.8712265491485596, + "learning_rate": 7.986509013545685e-07, + "loss": 0.1005, + "step": 381 + }, + { + "epoch": 0.4698646986469865, + "grad_norm": 3.7128748893737793, + "learning_rate": 7.391515966317919e-07, + "loss": 0.1948, + "step": 382 + }, + { + "epoch": 0.4710947109471095, + "grad_norm": 2.98445463180542, + "learning_rate": 6.81870231823969e-07, + "loss": 0.1618, + "step": 383 + }, + { + "epoch": 0.47232472324723246, + "grad_norm": 3.757805585861206, + "learning_rate": 6.268205256121396e-07, + "loss": 0.216, + "step": 384 + }, + { + "epoch": 0.4735547355473555, + "grad_norm": 2.943054676055908, + "learning_rate": 5.740156622031279e-07, + "loss": 0.132, + "step": 385 + }, + { + "epoch": 0.4747847478474785, + "grad_norm": 3.7545502185821533, + "learning_rate": 5.234682881719766e-07, + "loss": 0.2461, + "step": 386 + }, + { + "epoch": 0.47601476014760147, + "grad_norm": 3.329558849334717, + "learning_rate": 4.7519050943312215e-07, + "loss": 0.1825, + "step": 387 + }, + { + "epoch": 0.47724477244772445, + "grad_norm": 3.5239667892456055, + "learning_rate": 4.2919388834110066e-07, + "loss": 0.1912, + "step": 388 + }, + { + "epoch": 0.4784747847478475, + "grad_norm": 3.8842549324035645, + "learning_rate": 3.854894409213761e-07, + "loss": 0.1706, + "step": 389 + }, + { + "epoch": 0.4797047970479705, + "grad_norm": 4.085913181304932, + "learning_rate": 3.4408763423206093e-07, + "loss": 0.2599, + "step": 390 + }, + { + "epoch": 0.48093480934809346, + "grad_norm": 3.7395646572113037, + "learning_rate": 3.049983838570847e-07, + "loss": 0.1837, + "step": 391 + }, + { + "epoch": 0.4821648216482165, + "grad_norm": 4.188594818115234, + "learning_rate": 2.6823105153145125e-07, + "loss": 0.23, + "step": 392 + }, + { + "epoch": 0.4833948339483395, + "grad_norm": 3.445537805557251, + "learning_rate": 2.3379444289913233e-07, + "loss": 0.1523, + "step": 393 + }, + { + "epoch": 0.48462484624846247, + "grad_norm": 2.960714340209961, + "learning_rate": 2.016968054041546e-07, + "loss": 0.1105, + "step": 394 + }, + { + "epoch": 0.4858548585485855, + "grad_norm": 3.435288429260254, + "learning_rate": 1.7194582631535505e-07, + "loss": 0.1478, + "step": 395 + }, + { + "epoch": 0.4870848708487085, + "grad_norm": 3.486344814300537, + "learning_rate": 1.4454863088532388e-07, + "loss": 0.2184, + "step": 396 + }, + { + "epoch": 0.4883148831488315, + "grad_norm": 3.696117877960205, + "learning_rate": 1.19511780643915e-07, + "loss": 0.1977, + "step": 397 + }, + { + "epoch": 0.48954489544895446, + "grad_norm": 3.8439035415649414, + "learning_rate": 9.684127182679637e-08, + "loss": 0.1607, + "step": 398 + }, + { + "epoch": 0.4907749077490775, + "grad_norm": 3.716326951980591, + "learning_rate": 7.65425339393644e-08, + "loss": 0.2183, + "step": 399 + }, + { + "epoch": 0.4907749077490775, + "eval_loss": 0.10022835433483124, + "eval_runtime": 54.6223, + "eval_samples_per_second": 27.571, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8802787363695408, + "eval_sts-test_pearson_dot": 0.864146713903808, + "eval_sts-test_pearson_euclidean": 0.9095663542651595, + "eval_sts-test_pearson_manhattan": 0.9100280397532471, + "eval_sts-test_pearson_max": 0.9100280397532471, + "eval_sts-test_spearman_cosine": 0.9076901033093245, + "eval_sts-test_spearman_dot": 0.8705270000532351, + "eval_sts-test_spearman_euclidean": 0.9070381669333747, + "eval_sts-test_spearman_manhattan": 0.9074845160144926, + "eval_sts-test_spearman_max": 0.9076901033093245, + "step": 399 + }, + { + "epoch": 0.4920049200492005, + "grad_norm": 2.9559388160705566, + "learning_rate": 5.862042845640403e-08, + "loss": 0.1155, + "step": 400 + }, + { + "epoch": 0.4932349323493235, + "grad_norm": 4.184272289276123, + "learning_rate": 4.307924765777682e-08, + "loss": 0.2395, + "step": 401 + }, + { + "epoch": 0.4944649446494465, + "grad_norm": 3.0490658283233643, + "learning_rate": 2.992271360043986e-08, + "loss": 0.1194, + "step": 402 + }, + { + "epoch": 0.4956949569495695, + "grad_norm": 3.409501791000366, + "learning_rate": 1.9153977227022168e-08, + "loss": 0.1567, + "step": 403 + }, + { + "epoch": 0.4969249692496925, + "grad_norm": 2.8037118911743164, + "learning_rate": 1.0775617611189504e-08, + "loss": 0.1037, + "step": 404 + }, + { + "epoch": 0.4981549815498155, + "grad_norm": 3.936917781829834, + "learning_rate": 4.7896413399639575e-09, + "loss": 0.2713, + "step": 405 + }, + { + "epoch": 0.4993849938499385, + "grad_norm": 3.7185420989990234, + "learning_rate": 1.1974820331517313e-09, + "loss": 0.1742, + "step": 406 + }, + { + "epoch": 0.5006150061500615, + "grad_norm": 3.8419759273529053, + "learning_rate": 2e-05, + "loss": 0.221, + "step": 407 + }, + { + "epoch": 0.5018450184501845, + "grad_norm": 3.7907555103302, + "learning_rate": 1.9998802517966852e-05, + "loss": 0.1412, + "step": 408 + }, + { + "epoch": 0.5030750307503075, + "grad_norm": 3.7625885009765625, + "learning_rate": 1.9995210358660037e-05, + "loss": 0.1482, + "step": 409 + }, + { + "epoch": 0.5043050430504306, + "grad_norm": 3.234808921813965, + "learning_rate": 1.9989224382388813e-05, + "loss": 0.1347, + "step": 410 + }, + { + "epoch": 0.5055350553505535, + "grad_norm": 4.435244560241699, + "learning_rate": 1.9980846022772978e-05, + "loss": 0.2345, + "step": 411 + }, + { + "epoch": 0.5067650676506765, + "grad_norm": 3.1254684925079346, + "learning_rate": 1.997007728639956e-05, + "loss": 0.1231, + "step": 412 + }, + { + "epoch": 0.5079950799507995, + "grad_norm": 3.680490255355835, + "learning_rate": 1.9956920752342223e-05, + "loss": 0.1418, + "step": 413 + }, + { + "epoch": 0.5092250922509225, + "grad_norm": 3.7735965251922607, + "learning_rate": 1.9941379571543597e-05, + "loss": 0.152, + "step": 414 + }, + { + "epoch": 0.5104551045510455, + "grad_norm": 3.7166788578033447, + "learning_rate": 1.9923457466060637e-05, + "loss": 0.1878, + "step": 415 + }, + { + "epoch": 0.5116851168511685, + "grad_norm": 3.644469976425171, + "learning_rate": 1.9903158728173206e-05, + "loss": 0.1683, + "step": 416 + }, + { + "epoch": 0.5129151291512916, + "grad_norm": 3.0574779510498047, + "learning_rate": 1.9880488219356086e-05, + "loss": 0.1501, + "step": 417 + }, + { + "epoch": 0.5141451414514145, + "grad_norm": 4.234983444213867, + "learning_rate": 1.9855451369114677e-05, + "loss": 0.2589, + "step": 418 + }, + { + "epoch": 0.5153751537515375, + "grad_norm": 4.028163433074951, + "learning_rate": 1.9828054173684646e-05, + "loss": 0.1924, + "step": 419 + }, + { + "epoch": 0.5166051660516605, + "grad_norm": 2.9964284896850586, + "learning_rate": 1.9798303194595846e-05, + "loss": 0.1166, + "step": 420 + }, + { + "epoch": 0.5166051660516605, + "eval_loss": 0.09788215160369873, + "eval_runtime": 54.5712, + "eval_samples_per_second": 27.597, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8788928125426974, + "eval_sts-test_pearson_dot": 0.8655110850448609, + "eval_sts-test_pearson_euclidean": 0.9089126890484867, + "eval_sts-test_pearson_manhattan": 0.9096000817934198, + "eval_sts-test_pearson_max": 0.9096000817934198, + "eval_sts-test_spearman_cosine": 0.9078033687519063, + "eval_sts-test_spearman_dot": 0.8751717782243097, + "eval_sts-test_spearman_euclidean": 0.9071198827366754, + "eval_sts-test_spearman_manhattan": 0.9071902764708443, + "eval_sts-test_spearman_max": 0.9078033687519063, + "step": 420 + }, + { + "epoch": 0.5178351783517835, + "grad_norm": 3.0487561225891113, + "learning_rate": 1.9766205557100865e-05, + "loss": 0.1509, + "step": 421 + }, + { + "epoch": 0.5190651906519065, + "grad_norm": 3.460986375808716, + "learning_rate": 1.973176894846855e-05, + "loss": 0.1457, + "step": 422 + }, + { + "epoch": 0.5202952029520295, + "grad_norm": 3.831441879272461, + "learning_rate": 1.9695001616142916e-05, + "loss": 0.2244, + "step": 423 + }, + { + "epoch": 0.5215252152521526, + "grad_norm": 3.4312191009521484, + "learning_rate": 1.965591236576794e-05, + "loss": 0.1837, + "step": 424 + }, + { + "epoch": 0.5227552275522755, + "grad_norm": 4.465225696563721, + "learning_rate": 1.9614510559078622e-05, + "loss": 0.2649, + "step": 425 + }, + { + "epoch": 0.5239852398523985, + "grad_norm": 3.187695026397705, + "learning_rate": 1.95708061116589e-05, + "loss": 0.1295, + "step": 426 + }, + { + "epoch": 0.5252152521525215, + "grad_norm": 3.296201467514038, + "learning_rate": 1.952480949056688e-05, + "loss": 0.1776, + "step": 427 + }, + { + "epoch": 0.5264452644526445, + "grad_norm": 3.7875256538391113, + "learning_rate": 1.9476531711828027e-05, + "loss": 0.1949, + "step": 428 + }, + { + "epoch": 0.5276752767527675, + "grad_norm": 2.995544195175171, + "learning_rate": 1.942598433779687e-05, + "loss": 0.1262, + "step": 429 + }, + { + "epoch": 0.5289052890528906, + "grad_norm": 3.503070116043091, + "learning_rate": 1.9373179474387858e-05, + "loss": 0.1502, + "step": 430 + }, + { + "epoch": 0.5301353013530136, + "grad_norm": 3.6476597785949707, + "learning_rate": 1.9318129768176036e-05, + "loss": 0.1927, + "step": 431 + }, + { + "epoch": 0.5313653136531366, + "grad_norm": 3.9136264324188232, + "learning_rate": 1.926084840336821e-05, + "loss": 0.2161, + "step": 432 + }, + { + "epoch": 0.5325953259532595, + "grad_norm": 3.655367374420166, + "learning_rate": 1.9201349098645433e-05, + "loss": 0.2082, + "step": 433 + }, + { + "epoch": 0.5338253382533825, + "grad_norm": 3.711770534515381, + "learning_rate": 1.9139646103877378e-05, + "loss": 0.2171, + "step": 434 + }, + { + "epoch": 0.5350553505535055, + "grad_norm": 3.502131938934326, + "learning_rate": 1.9075754196709574e-05, + "loss": 0.209, + "step": 435 + }, + { + "epoch": 0.5362853628536285, + "grad_norm": 3.669158935546875, + "learning_rate": 1.9009688679024195e-05, + "loss": 0.1841, + "step": 436 + }, + { + "epoch": 0.5375153751537516, + "grad_norm": 3.3147544860839844, + "learning_rate": 1.894146537327533e-05, + "loss": 0.1522, + "step": 437 + }, + { + "epoch": 0.5387453874538746, + "grad_norm": 3.6316304206848145, + "learning_rate": 1.887110061869956e-05, + "loss": 0.1644, + "step": 438 + }, + { + "epoch": 0.5399753997539976, + "grad_norm": 3.4334845542907715, + "learning_rate": 1.8798611267402748e-05, + "loss": 0.1784, + "step": 439 + }, + { + "epoch": 0.5412054120541205, + "grad_norm": 3.5507333278656006, + "learning_rate": 1.8724014680324057e-05, + "loss": 0.2041, + "step": 440 + }, + { + "epoch": 0.5424354243542435, + "grad_norm": 3.303339719772339, + "learning_rate": 1.864732872307804e-05, + "loss": 0.1564, + "step": 441 + }, + { + "epoch": 0.5424354243542435, + "eval_loss": 0.09677440673112869, + "eval_runtime": 54.7177, + "eval_samples_per_second": 27.523, + "eval_steps_per_second": 0.219, + "eval_sts-test_pearson_cosine": 0.8770540569686203, + "eval_sts-test_pearson_dot": 0.8623917149869358, + "eval_sts-test_pearson_euclidean": 0.9077935185844548, + "eval_sts-test_pearson_manhattan": 0.9085145036943166, + "eval_sts-test_pearson_max": 0.9085145036943166, + "eval_sts-test_spearman_cosine": 0.9058137276704445, + "eval_sts-test_spearman_dot": 0.8694433482586422, + "eval_sts-test_spearman_euclidean": 0.9053487173133266, + "eval_sts-test_spearman_manhattan": 0.9059988188875594, + "eval_sts-test_spearman_max": 0.9059988188875594, + "step": 441 + }, + { + "epoch": 0.5436654366543665, + "grad_norm": 3.936544179916382, + "learning_rate": 1.8568571761675893e-05, + "loss": 0.2151, + "step": 442 + }, + { + "epoch": 0.5448954489544895, + "grad_norm": 3.5802316665649414, + "learning_rate": 1.8487762658126872e-05, + "loss": 0.1797, + "step": 443 + }, + { + "epoch": 0.5461254612546126, + "grad_norm": 3.9694156646728516, + "learning_rate": 1.8404920765920895e-05, + "loss": 0.1652, + "step": 444 + }, + { + "epoch": 0.5473554735547356, + "grad_norm": 3.010603666305542, + "learning_rate": 1.8320065925393468e-05, + "loss": 0.1561, + "step": 445 + }, + { + "epoch": 0.5485854858548586, + "grad_norm": 2.738050699234009, + "learning_rate": 1.8233218458973984e-05, + "loss": 0.1063, + "step": 446 + }, + { + "epoch": 0.5498154981549815, + "grad_norm": 3.3592913150787354, + "learning_rate": 1.814439916631857e-05, + "loss": 0.1584, + "step": 447 + }, + { + "epoch": 0.5510455104551045, + "grad_norm": 3.9499318599700928, + "learning_rate": 1.805362931932866e-05, + "loss": 0.2396, + "step": 448 + }, + { + "epoch": 0.5522755227552275, + "grad_norm": 3.8009963035583496, + "learning_rate": 1.796093065705644e-05, + "loss": 0.1952, + "step": 449 + }, + { + "epoch": 0.5535055350553506, + "grad_norm": 3.4526023864746094, + "learning_rate": 1.786632538049842e-05, + "loss": 0.1598, + "step": 450 + }, + { + "epoch": 0.5547355473554736, + "grad_norm": 3.8199033737182617, + "learning_rate": 1.7769836147278385e-05, + "loss": 0.2093, + "step": 451 + }, + { + "epoch": 0.5559655596555966, + "grad_norm": 3.7011399269104004, + "learning_rate": 1.7671486066220965e-05, + "loss": 0.1585, + "step": 452 + }, + { + "epoch": 0.5571955719557196, + "grad_norm": 4.342462539672852, + "learning_rate": 1.757129869181718e-05, + "loss": 0.2311, + "step": 453 + }, + { + "epoch": 0.5584255842558425, + "grad_norm": 2.5770063400268555, + "learning_rate": 1.746929801858317e-05, + "loss": 0.1048, + "step": 454 + }, + { + "epoch": 0.5596555965559655, + "grad_norm": 3.490851640701294, + "learning_rate": 1.7365508475313663e-05, + "loss": 0.1571, + "step": 455 + }, + { + "epoch": 0.5608856088560885, + "grad_norm": 3.9881510734558105, + "learning_rate": 1.725995491923131e-05, + "loss": 0.1915, + "step": 456 + }, + { + "epoch": 0.5621156211562116, + "grad_norm": 3.580413579940796, + "learning_rate": 1.7152662630033506e-05, + "loss": 0.1625, + "step": 457 + }, + { + "epoch": 0.5633456334563346, + "grad_norm": 3.502101182937622, + "learning_rate": 1.7043657303837965e-05, + "loss": 0.1613, + "step": 458 + }, + { + "epoch": 0.5645756457564576, + "grad_norm": 3.486637830734253, + "learning_rate": 1.6932965047028624e-05, + "loss": 0.1845, + "step": 459 + }, + { + "epoch": 0.5658056580565806, + "grad_norm": 4.057269096374512, + "learning_rate": 1.682061237000322e-05, + "loss": 0.2134, + "step": 460 + }, + { + "epoch": 0.5670356703567035, + "grad_norm": 3.6247904300689697, + "learning_rate": 1.670662618082419e-05, + "loss": 0.2059, + "step": 461 + }, + { + "epoch": 0.5682656826568265, + "grad_norm": 3.409250497817993, + "learning_rate": 1.659103377877423e-05, + "loss": 0.1974, + "step": 462 + }, + { + "epoch": 0.5682656826568265, + "eval_loss": 0.09474142640829086, + "eval_runtime": 54.6525, + "eval_samples_per_second": 27.556, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8787555052231211, + "eval_sts-test_pearson_dot": 0.8667506982700097, + "eval_sts-test_pearson_euclidean": 0.9084646461700655, + "eval_sts-test_pearson_manhattan": 0.9091836932503693, + "eval_sts-test_pearson_max": 0.9091836932503693, + "eval_sts-test_spearman_cosine": 0.9067327395743786, + "eval_sts-test_spearman_dot": 0.8779231305259144, + "eval_sts-test_spearman_euclidean": 0.9060663037897377, + "eval_sts-test_spearman_manhattan": 0.9065003462739729, + "eval_sts-test_spearman_max": 0.9067327395743786, + "step": 462 + }, + { + "epoch": 0.5694956949569495, + "grad_norm": 3.437278985977173, + "learning_rate": 1.647386284781828e-05, + "loss": 0.1624, + "step": 463 + }, + { + "epoch": 0.5707257072570726, + "grad_norm": 3.9308578968048096, + "learning_rate": 1.6355141449973254e-05, + "loss": 0.2005, + "step": 464 + }, + { + "epoch": 0.5719557195571956, + "grad_norm": 3.612074375152588, + "learning_rate": 1.623489801858734e-05, + "loss": 0.1407, + "step": 465 + }, + { + "epoch": 0.5731857318573186, + "grad_norm": 3.1348252296447754, + "learning_rate": 1.611316135153026e-05, + "loss": 0.1175, + "step": 466 + }, + { + "epoch": 0.5744157441574416, + "grad_norm": 3.639906883239746, + "learning_rate": 1.5989960604296335e-05, + "loss": 0.1888, + "step": 467 + }, + { + "epoch": 0.5756457564575646, + "grad_norm": 3.182251453399658, + "learning_rate": 1.5865325283021826e-05, + "loss": 0.1423, + "step": 468 + }, + { + "epoch": 0.5768757687576875, + "grad_norm": 2.601531505584717, + "learning_rate": 1.573928523741832e-05, + "loss": 0.1195, + "step": 469 + }, + { + "epoch": 0.5781057810578106, + "grad_norm": 3.6461949348449707, + "learning_rate": 1.5611870653623826e-05, + "loss": 0.1525, + "step": 470 + }, + { + "epoch": 0.5793357933579336, + "grad_norm": 4.034285545349121, + "learning_rate": 1.5483112046973307e-05, + "loss": 0.2155, + "step": 471 + }, + { + "epoch": 0.5805658056580566, + "grad_norm": 3.840399980545044, + "learning_rate": 1.535304025469039e-05, + "loss": 0.2048, + "step": 472 + }, + { + "epoch": 0.5817958179581796, + "grad_norm": 4.369460105895996, + "learning_rate": 1.5221686428501929e-05, + "loss": 0.2386, + "step": 473 + }, + { + "epoch": 0.5830258302583026, + "grad_norm": 3.9414284229278564, + "learning_rate": 1.5089082027177293e-05, + "loss": 0.162, + "step": 474 + }, + { + "epoch": 0.5842558425584256, + "grad_norm": 3.5878100395202637, + "learning_rate": 1.4955258808994093e-05, + "loss": 0.1735, + "step": 475 + }, + { + "epoch": 0.5854858548585485, + "grad_norm": 3.7590479850769043, + "learning_rate": 1.4820248824132218e-05, + "loss": 0.2067, + "step": 476 + }, + { + "epoch": 0.5867158671586716, + "grad_norm": 3.3892862796783447, + "learning_rate": 1.4684084406997903e-05, + "loss": 0.1395, + "step": 477 + }, + { + "epoch": 0.5879458794587946, + "grad_norm": 3.4751815795898438, + "learning_rate": 1.4546798168479758e-05, + "loss": 0.1482, + "step": 478 + }, + { + "epoch": 0.5891758917589176, + "grad_norm": 4.276810169219971, + "learning_rate": 1.4408422988138585e-05, + "loss": 0.2399, + "step": 479 + }, + { + "epoch": 0.5904059040590406, + "grad_norm": 3.408428192138672, + "learning_rate": 1.4268992006332845e-05, + "loss": 0.1849, + "step": 480 + }, + { + "epoch": 0.5916359163591636, + "grad_norm": 3.2603816986083984, + "learning_rate": 1.412853861628166e-05, + "loss": 0.139, + "step": 481 + }, + { + "epoch": 0.5928659286592866, + "grad_norm": 4.1905975341796875, + "learning_rate": 1.398709645606724e-05, + "loss": 0.2089, + "step": 482 + }, + { + "epoch": 0.5940959409594095, + "grad_norm": 4.082717418670654, + "learning_rate": 1.3844699400578696e-05, + "loss": 0.2066, + "step": 483 + }, + { + "epoch": 0.5940959409594095, + "eval_loss": 0.09337170422077179, + "eval_runtime": 54.624, + "eval_samples_per_second": 27.57, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8777835156577275, + "eval_sts-test_pearson_dot": 0.8618993321533206, + "eval_sts-test_pearson_euclidean": 0.90904681111194, + "eval_sts-test_pearson_manhattan": 0.9093931766288231, + "eval_sts-test_pearson_max": 0.9093931766288231, + "eval_sts-test_spearman_cosine": 0.9072326111641204, + "eval_sts-test_spearman_dot": 0.8735216649630247, + "eval_sts-test_spearman_euclidean": 0.9072676066505284, + "eval_sts-test_spearman_manhattan": 0.9076947126892481, + "eval_sts-test_spearman_max": 0.9076947126892481, + "step": 483 + }, + { + "epoch": 0.5953259532595326, + "grad_norm": 4.576707363128662, + "learning_rate": 1.3701381553399142e-05, + "loss": 0.2293, + "step": 484 + }, + { + "epoch": 0.5965559655596556, + "grad_norm": 3.751978635787964, + "learning_rate": 1.3557177238637987e-05, + "loss": 0.1919, + "step": 485 + }, + { + "epoch": 0.5977859778597786, + "grad_norm": 3.1672027111053467, + "learning_rate": 1.3412120992710432e-05, + "loss": 0.1168, + "step": 486 + }, + { + "epoch": 0.5990159901599016, + "grad_norm": 3.9741456508636475, + "learning_rate": 1.3266247556066122e-05, + "loss": 0.2057, + "step": 487 + }, + { + "epoch": 0.6002460024600246, + "grad_norm": 3.6646575927734375, + "learning_rate": 1.3119591864868977e-05, + "loss": 0.1866, + "step": 488 + }, + { + "epoch": 0.6014760147601476, + "grad_norm": 3.928297519683838, + "learning_rate": 1.2972189042630046e-05, + "loss": 0.2277, + "step": 489 + }, + { + "epoch": 0.6027060270602707, + "grad_norm": 3.5202107429504395, + "learning_rate": 1.2824074391795578e-05, + "loss": 0.1527, + "step": 490 + }, + { + "epoch": 0.6039360393603936, + "grad_norm": 4.187824726104736, + "learning_rate": 1.2675283385292212e-05, + "loss": 0.275, + "step": 491 + }, + { + "epoch": 0.6051660516605166, + "grad_norm": 3.150355815887451, + "learning_rate": 1.2525851658031349e-05, + "loss": 0.1212, + "step": 492 + }, + { + "epoch": 0.6063960639606396, + "grad_norm": 3.475013256072998, + "learning_rate": 1.2375814998374715e-05, + "loss": 0.1384, + "step": 493 + }, + { + "epoch": 0.6076260762607626, + "grad_norm": 3.4592478275299072, + "learning_rate": 1.2225209339563153e-05, + "loss": 0.1611, + "step": 494 + }, + { + "epoch": 0.6088560885608856, + "grad_norm": 3.5501179695129395, + "learning_rate": 1.2074070751110747e-05, + "loss": 0.145, + "step": 495 + }, + { + "epoch": 0.6100861008610086, + "grad_norm": 3.7718143463134766, + "learning_rate": 1.192243543016637e-05, + "loss": 0.1996, + "step": 496 + }, + { + "epoch": 0.6113161131611317, + "grad_norm": 4.389284133911133, + "learning_rate": 1.1770339692844482e-05, + "loss": 0.3, + "step": 497 + }, + { + "epoch": 0.6125461254612546, + "grad_norm": 3.2603354454040527, + "learning_rate": 1.161781996552765e-05, + "loss": 0.1117, + "step": 498 + }, + { + "epoch": 0.6137761377613776, + "grad_norm": 3.7046337127685547, + "learning_rate": 1.1464912776142489e-05, + "loss": 0.1905, + "step": 499 + }, + { + "epoch": 0.6150061500615006, + "grad_norm": 4.1772027015686035, + "learning_rate": 1.1311654745411424e-05, + "loss": 0.2221, + "step": 500 + }, + { + "epoch": 0.6162361623616236, + "grad_norm": 3.1302425861358643, + "learning_rate": 1.1158082578082089e-05, + "loss": 0.1749, + "step": 501 + }, + { + "epoch": 0.6174661746617466, + "grad_norm": 3.4048593044281006, + "learning_rate": 1.1004233054136726e-05, + "loss": 0.1533, + "step": 502 + }, + { + "epoch": 0.6186961869618696, + "grad_norm": 3.637916088104248, + "learning_rate": 1.085014301998347e-05, + "loss": 0.2268, + "step": 503 + }, + { + "epoch": 0.6199261992619927, + "grad_norm": 3.434365749359131, + "learning_rate": 1.0695849379631816e-05, + "loss": 0.1879, + "step": 504 + }, + { + "epoch": 0.6199261992619927, + "eval_loss": 0.09355636686086655, + "eval_runtime": 54.5989, + "eval_samples_per_second": 27.583, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8776714319316713, + "eval_sts-test_pearson_dot": 0.8624508191789559, + "eval_sts-test_pearson_euclidean": 0.9088025939622317, + "eval_sts-test_pearson_manhattan": 0.9092112976264597, + "eval_sts-test_pearson_max": 0.9092112976264597, + "eval_sts-test_spearman_cosine": 0.9066244415218112, + "eval_sts-test_spearman_dot": 0.8720764677276509, + "eval_sts-test_spearman_euclidean": 0.9063792941119092, + "eval_sts-test_spearman_manhattan": 0.9068959921759343, + "eval_sts-test_spearman_max": 0.9068959921759343, + "step": 504 + }, + { + "epoch": 0.6211562115621156, + "grad_norm": 4.41910982131958, + "learning_rate": 1.0541389085854174e-05, + "loss": 0.2956, + "step": 505 + }, + { + "epoch": 0.6223862238622386, + "grad_norm": 3.4075140953063965, + "learning_rate": 1.038679913133589e-05, + "loss": 0.1566, + "step": 506 + }, + { + "epoch": 0.6236162361623616, + "grad_norm": 3.034949779510498, + "learning_rate": 1.0232116539815556e-05, + "loss": 0.1612, + "step": 507 + }, + { + "epoch": 0.6248462484624846, + "grad_norm": 4.093225479125977, + "learning_rate": 1.0077378357218023e-05, + "loss": 0.2312, + "step": 508 + }, + { + "epoch": 0.6260762607626076, + "grad_norm": 3.869215250015259, + "learning_rate": 9.92262164278198e-06, + "loss": 0.181, + "step": 509 + }, + { + "epoch": 0.6273062730627307, + "grad_norm": 4.116591930389404, + "learning_rate": 9.76788346018444e-06, + "loss": 0.235, + "step": 510 + }, + { + "epoch": 0.6285362853628537, + "grad_norm": 3.339435577392578, + "learning_rate": 9.613200868664105e-06, + "loss": 0.1376, + "step": 511 + }, + { + "epoch": 0.6297662976629766, + "grad_norm": 2.6395070552825928, + "learning_rate": 9.458610914145821e-06, + "loss": 0.1066, + "step": 512 + }, + { + "epoch": 0.6309963099630996, + "grad_norm": 3.9118542671203613, + "learning_rate": 9.304150620368189e-06, + "loss": 0.2235, + "step": 513 + }, + { + "epoch": 0.6322263222632226, + "grad_norm": 4.029881000518799, + "learning_rate": 9.149856980016527e-06, + "loss": 0.2549, + "step": 514 + }, + { + "epoch": 0.6334563345633456, + "grad_norm": 4.164516925811768, + "learning_rate": 8.99576694586327e-06, + "loss": 0.2676, + "step": 515 + }, + { + "epoch": 0.6346863468634686, + "grad_norm": 3.3640923500061035, + "learning_rate": 8.841917421917908e-06, + "loss": 0.1652, + "step": 516 + }, + { + "epoch": 0.6359163591635917, + "grad_norm": 3.2747318744659424, + "learning_rate": 8.688345254588579e-06, + "loss": 0.1573, + "step": 517 + }, + { + "epoch": 0.6371463714637147, + "grad_norm": 3.6047604084014893, + "learning_rate": 8.535087223857506e-06, + "loss": 0.2106, + "step": 518 + }, + { + "epoch": 0.6383763837638377, + "grad_norm": 3.269946336746216, + "learning_rate": 8.382180034472346e-06, + "loss": 0.151, + "step": 519 + }, + { + "epoch": 0.6396063960639606, + "grad_norm": 3.2695987224578857, + "learning_rate": 8.229660307155513e-06, + "loss": 0.1491, + "step": 520 + }, + { + "epoch": 0.6408364083640836, + "grad_norm": 4.302383899688721, + "learning_rate": 8.077564569833633e-06, + "loss": 0.2612, + "step": 521 + }, + { + "epoch": 0.6420664206642066, + "grad_norm": 3.2373013496398926, + "learning_rate": 7.92592924888925e-06, + "loss": 0.1287, + "step": 522 + }, + { + "epoch": 0.6432964329643296, + "grad_norm": 4.335332870483398, + "learning_rate": 7.77479066043685e-06, + "loss": 0.2084, + "step": 523 + }, + { + "epoch": 0.6445264452644527, + "grad_norm": 3.5337517261505127, + "learning_rate": 7.624185001625287e-06, + "loss": 0.1545, + "step": 524 + }, + { + "epoch": 0.6457564575645757, + "grad_norm": 3.512129783630371, + "learning_rate": 7.4741483419686525e-06, + "loss": 0.1946, + "step": 525 + }, + { + "epoch": 0.6457564575645757, + "eval_loss": 0.09311617165803909, + "eval_runtime": 54.6375, + "eval_samples_per_second": 27.564, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.877050544612564, + "eval_sts-test_pearson_dot": 0.8622944405839497, + "eval_sts-test_pearson_euclidean": 0.9083626014354534, + "eval_sts-test_pearson_manhattan": 0.9088804166416584, + "eval_sts-test_pearson_max": 0.9088804166416584, + "eval_sts-test_spearman_cosine": 0.9061299400834363, + "eval_sts-test_spearman_dot": 0.8716984538226776, + "eval_sts-test_spearman_euclidean": 0.9057804327319695, + "eval_sts-test_spearman_manhattan": 0.9065156959566302, + "eval_sts-test_spearman_max": 0.9065156959566302, + "step": 525 + }, + { + "epoch": 0.6469864698646987, + "grad_norm": 3.5218262672424316, + "learning_rate": 7.324716614707792e-06, + "loss": 0.1684, + "step": 526 + }, + { + "epoch": 0.6482164821648216, + "grad_norm": 3.742743968963623, + "learning_rate": 7.175925608204424e-06, + "loss": 0.1974, + "step": 527 + }, + { + "epoch": 0.6494464944649446, + "grad_norm": 3.802528142929077, + "learning_rate": 7.0278109573699574e-06, + "loss": 0.2448, + "step": 528 + }, + { + "epoch": 0.6506765067650676, + "grad_norm": 4.072224140167236, + "learning_rate": 6.880408135131027e-06, + "loss": 0.2255, + "step": 529 + }, + { + "epoch": 0.6519065190651907, + "grad_norm": 3.6978464126586914, + "learning_rate": 6.733752443933879e-06, + "loss": 0.2157, + "step": 530 + }, + { + "epoch": 0.6531365313653137, + "grad_norm": 3.507934808731079, + "learning_rate": 6.587879007289572e-06, + "loss": 0.1948, + "step": 531 + }, + { + "epoch": 0.6543665436654367, + "grad_norm": 3.386146306991577, + "learning_rate": 6.442822761362015e-06, + "loss": 0.1418, + "step": 532 + }, + { + "epoch": 0.6555965559655597, + "grad_norm": 3.5481231212615967, + "learning_rate": 6.29861844660086e-06, + "loss": 0.1683, + "step": 533 + }, + { + "epoch": 0.6568265682656826, + "grad_norm": 3.780062437057495, + "learning_rate": 6.155300599421305e-06, + "loss": 0.193, + "step": 534 + }, + { + "epoch": 0.6580565805658056, + "grad_norm": 4.013569355010986, + "learning_rate": 6.012903543932762e-06, + "loss": 0.2341, + "step": 535 + }, + { + "epoch": 0.6592865928659286, + "grad_norm": 2.8476858139038086, + "learning_rate": 5.871461383718344e-06, + "loss": 0.131, + "step": 536 + }, + { + "epoch": 0.6605166051660517, + "grad_norm": 3.4684088230133057, + "learning_rate": 5.731007993667159e-06, + "loss": 0.1733, + "step": 537 + }, + { + "epoch": 0.6617466174661747, + "grad_norm": 3.539458751678467, + "learning_rate": 5.5915770118614195e-06, + "loss": 0.1489, + "step": 538 + }, + { + "epoch": 0.6629766297662977, + "grad_norm": 3.6291494369506836, + "learning_rate": 5.453201831520245e-06, + "loss": 0.1918, + "step": 539 + }, + { + "epoch": 0.6642066420664207, + "grad_norm": 3.818389415740967, + "learning_rate": 5.3159155930021e-06, + "loss": 0.1953, + "step": 540 + }, + { + "epoch": 0.6654366543665436, + "grad_norm": 3.375427007675171, + "learning_rate": 5.179751175867787e-06, + "loss": 0.1421, + "step": 541 + }, + { + "epoch": 0.6666666666666666, + "grad_norm": 3.8009254932403564, + "learning_rate": 5.044741191005908e-06, + "loss": 0.2214, + "step": 542 + }, + { + "epoch": 0.6678966789667896, + "grad_norm": 3.446690082550049, + "learning_rate": 4.91091797282271e-06, + "loss": 0.2152, + "step": 543 + }, + { + "epoch": 0.6691266912669127, + "grad_norm": 3.6333260536193848, + "learning_rate": 4.778313571498075e-06, + "loss": 0.209, + "step": 544 + }, + { + "epoch": 0.6703567035670357, + "grad_norm": 3.6350371837615967, + "learning_rate": 4.646959745309614e-06, + "loss": 0.1735, + "step": 545 + }, + { + "epoch": 0.6715867158671587, + "grad_norm": 3.8228209018707275, + "learning_rate": 4.516887953026694e-06, + "loss": 0.2048, + "step": 546 + }, + { + "epoch": 0.6715867158671587, + "eval_loss": 0.09180988371372223, + "eval_runtime": 54.6293, + "eval_samples_per_second": 27.568, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8779366163759763, + "eval_sts-test_pearson_dot": 0.8633199514924738, + "eval_sts-test_pearson_euclidean": 0.908752170172483, + "eval_sts-test_pearson_manhattan": 0.9093094000687254, + "eval_sts-test_pearson_max": 0.9093094000687254, + "eval_sts-test_spearman_cosine": 0.9060224923048356, + "eval_sts-test_spearman_dot": 0.8726965859567515, + "eval_sts-test_spearman_euclidean": 0.9059672692482788, + "eval_sts-test_spearman_manhattan": 0.9067383334820525, + "eval_sts-test_spearman_max": 0.9067383334820525, + "step": 546 + }, + { + "epoch": 0.6728167281672817, + "grad_norm": 3.953333854675293, + "learning_rate": 4.388129346376177e-06, + "loss": 0.1721, + "step": 547 + }, + { + "epoch": 0.6740467404674046, + "grad_norm": 3.85856294631958, + "learning_rate": 4.260714762581682e-06, + "loss": 0.1838, + "step": 548 + }, + { + "epoch": 0.6752767527675276, + "grad_norm": 3.620084047317505, + "learning_rate": 4.134674716978176e-06, + "loss": 0.1614, + "step": 549 + }, + { + "epoch": 0.6765067650676507, + "grad_norm": 3.3145596981048584, + "learning_rate": 4.010039395703665e-06, + "loss": 0.1999, + "step": 550 + }, + { + "epoch": 0.6777367773677737, + "grad_norm": 2.761929988861084, + "learning_rate": 3.886838648469742e-06, + "loss": 0.0984, + "step": 551 + }, + { + "epoch": 0.6789667896678967, + "grad_norm": 2.96940279006958, + "learning_rate": 3.765101981412669e-06, + "loss": 0.1351, + "step": 552 + }, + { + "epoch": 0.6801968019680197, + "grad_norm": 3.704197645187378, + "learning_rate": 3.644858550026752e-06, + "loss": 0.1886, + "step": 553 + }, + { + "epoch": 0.6814268142681427, + "grad_norm": 2.9747583866119385, + "learning_rate": 3.5261371521817276e-06, + "loss": 0.1148, + "step": 554 + }, + { + "epoch": 0.6826568265682657, + "grad_norm": 3.257741689682007, + "learning_rate": 3.4089662212257733e-06, + "loss": 0.1766, + "step": 555 + }, + { + "epoch": 0.6838868388683886, + "grad_norm": 3.7708511352539062, + "learning_rate": 3.293373819175819e-06, + "loss": 0.19, + "step": 556 + }, + { + "epoch": 0.6851168511685117, + "grad_norm": 3.6153717041015625, + "learning_rate": 3.1793876299967874e-06, + "loss": 0.2082, + "step": 557 + }, + { + "epoch": 0.6863468634686347, + "grad_norm": 3.6002707481384277, + "learning_rate": 3.067034952971382e-06, + "loss": 0.222, + "step": 558 + }, + { + "epoch": 0.6875768757687577, + "grad_norm": 3.7031867504119873, + "learning_rate": 2.956342696162037e-06, + "loss": 0.2032, + "step": 559 + }, + { + "epoch": 0.6888068880688807, + "grad_norm": 3.727587938308716, + "learning_rate": 2.847337369966502e-06, + "loss": 0.1854, + "step": 560 + }, + { + "epoch": 0.6900369003690037, + "grad_norm": 2.8371074199676514, + "learning_rate": 2.7400450807686973e-06, + "loss": 0.1473, + "step": 561 + }, + { + "epoch": 0.6912669126691267, + "grad_norm": 4.017850875854492, + "learning_rate": 2.6344915246863445e-06, + "loss": 0.2003, + "step": 562 + }, + { + "epoch": 0.6924969249692496, + "grad_norm": 3.079763174057007, + "learning_rate": 2.5307019814168343e-06, + "loss": 0.1223, + "step": 563 + }, + { + "epoch": 0.6937269372693727, + "grad_norm": 4.067211151123047, + "learning_rate": 2.428701308182826e-06, + "loss": 0.2319, + "step": 564 + }, + { + "epoch": 0.6949569495694957, + "grad_norm": 2.728217363357544, + "learning_rate": 2.328513933779031e-06, + "loss": 0.0761, + "step": 565 + }, + { + "epoch": 0.6961869618696187, + "grad_norm": 4.238541126251221, + "learning_rate": 2.230163852721614e-06, + "loss": 0.2835, + "step": 566 + }, + { + "epoch": 0.6974169741697417, + "grad_norm": 3.9363412857055664, + "learning_rate": 2.1336746195015824e-06, + "loss": 0.2331, + "step": 567 + }, + { + "epoch": 0.6974169741697417, + "eval_loss": 0.09195715188980103, + "eval_runtime": 54.5928, + "eval_samples_per_second": 27.586, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8779204832859013, + "eval_sts-test_pearson_dot": 0.8640417892863237, + "eval_sts-test_pearson_euclidean": 0.9085408480141077, + "eval_sts-test_pearson_manhattan": 0.9091906262189161, + "eval_sts-test_pearson_max": 0.9091906262189161, + "eval_sts-test_spearman_cosine": 0.9061162909487119, + "eval_sts-test_spearman_dot": 0.8733624399749941, + "eval_sts-test_spearman_euclidean": 0.9060925280289132, + "eval_sts-test_spearman_manhattan": 0.9065397721352588, + "eval_sts-test_spearman_max": 0.9065397721352588, + "step": 567 + }, + { + "epoch": 0.6986469864698647, + "grad_norm": 3.543086051940918, + "learning_rate": 2.0390693429435626e-06, + "loss": 0.1698, + "step": 568 + }, + { + "epoch": 0.6998769987699877, + "grad_norm": 4.054234981536865, + "learning_rate": 1.946370680671339e-06, + "loss": 0.203, + "step": 569 + }, + { + "epoch": 0.7011070110701108, + "grad_norm": 4.240468502044678, + "learning_rate": 1.8556008336814268e-06, + "loss": 0.2344, + "step": 570 + }, + { + "epoch": 0.7023370233702337, + "grad_norm": 3.820857524871826, + "learning_rate": 1.7667815410260147e-06, + "loss": 0.1823, + "step": 571 + }, + { + "epoch": 0.7035670356703567, + "grad_norm": 3.1917405128479004, + "learning_rate": 1.679934074606533e-06, + "loss": 0.2043, + "step": 572 + }, + { + "epoch": 0.7047970479704797, + "grad_norm": 4.044501304626465, + "learning_rate": 1.5950792340791044e-06, + "loss": 0.1881, + "step": 573 + }, + { + "epoch": 0.7060270602706027, + "grad_norm": 3.4755239486694336, + "learning_rate": 1.5122373418731262e-06, + "loss": 0.1599, + "step": 574 + }, + { + "epoch": 0.7072570725707257, + "grad_norm": 3.088360071182251, + "learning_rate": 1.4314282383241073e-06, + "loss": 0.0829, + "step": 575 + }, + { + "epoch": 0.7084870848708487, + "grad_norm": 3.2899420261383057, + "learning_rate": 1.3526712769219619e-06, + "loss": 0.1816, + "step": 576 + }, + { + "epoch": 0.7097170971709718, + "grad_norm": 4.0722336769104, + "learning_rate": 1.275985319675943e-06, + "loss": 0.1801, + "step": 577 + }, + { + "epoch": 0.7109471094710947, + "grad_norm": 3.4605586528778076, + "learning_rate": 1.201388732597253e-06, + "loss": 0.1707, + "step": 578 + }, + { + "epoch": 0.7121771217712177, + "grad_norm": 3.503009796142578, + "learning_rate": 1.1288993813004446e-06, + "loss": 0.2306, + "step": 579 + }, + { + "epoch": 0.7134071340713407, + "grad_norm": 3.4009296894073486, + "learning_rate": 1.0585346267246743e-06, + "loss": 0.1503, + "step": 580 + }, + { + "epoch": 0.7146371463714637, + "grad_norm": 3.8025364875793457, + "learning_rate": 9.903113209758098e-07, + "loss": 0.1779, + "step": 581 + }, + { + "epoch": 0.7158671586715867, + "grad_norm": 3.1331369876861572, + "learning_rate": 9.242458032904266e-07, + "loss": 0.1422, + "step": 582 + }, + { + "epoch": 0.7170971709717097, + "grad_norm": 3.1663196086883545, + "learning_rate": 8.603538961226233e-07, + "loss": 0.1358, + "step": 583 + }, + { + "epoch": 0.7183271832718328, + "grad_norm": 2.648996591567993, + "learning_rate": 7.986509013545685e-07, + "loss": 0.0978, + "step": 584 + }, + { + "epoch": 0.7195571955719557, + "grad_norm": 3.846203088760376, + "learning_rate": 7.391515966317908e-07, + "loss": 0.1713, + "step": 585 + }, + { + "epoch": 0.7207872078720787, + "grad_norm": 3.713327169418335, + "learning_rate": 6.818702318239678e-07, + "loss": 0.1771, + "step": 586 + }, + { + "epoch": 0.7220172201722017, + "grad_norm": 3.2118642330169678, + "learning_rate": 6.268205256121396e-07, + "loss": 0.1241, + "step": 587 + }, + { + "epoch": 0.7232472324723247, + "grad_norm": 3.1194214820861816, + "learning_rate": 5.740156622031312e-07, + "loss": 0.1267, + "step": 588 + }, + { + "epoch": 0.7232472324723247, + "eval_loss": 0.09176090359687805, + "eval_runtime": 54.6105, + "eval_samples_per_second": 27.577, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8782033315155763, + "eval_sts-test_pearson_dot": 0.86482127981954, + "eval_sts-test_pearson_euclidean": 0.9087286707257626, + "eval_sts-test_pearson_manhattan": 0.9093761455537492, + "eval_sts-test_pearson_max": 0.9093761455537492, + "eval_sts-test_spearman_cosine": 0.9063635864191608, + "eval_sts-test_spearman_dot": 0.8744101117554283, + "eval_sts-test_spearman_euclidean": 0.9063503400457888, + "eval_sts-test_spearman_manhattan": 0.9067913189755399, + "eval_sts-test_spearman_max": 0.9067913189755399, + "step": 588 + }, + { + "epoch": 0.7244772447724477, + "grad_norm": 3.13580584526062, + "learning_rate": 5.234682881719766e-07, + "loss": 0.1126, + "step": 589 + }, + { + "epoch": 0.7257072570725708, + "grad_norm": 2.457448959350586, + "learning_rate": 4.7519050943312215e-07, + "loss": 0.0858, + "step": 590 + }, + { + "epoch": 0.7269372693726938, + "grad_norm": 2.800050973892212, + "learning_rate": 4.2919388834110066e-07, + "loss": 0.1335, + "step": 591 + }, + { + "epoch": 0.7281672816728167, + "grad_norm": 3.386810541152954, + "learning_rate": 3.854894409213772e-07, + "loss": 0.1958, + "step": 592 + }, + { + "epoch": 0.7293972939729397, + "grad_norm": 3.494534492492676, + "learning_rate": 3.4408763423206093e-07, + "loss": 0.1448, + "step": 593 + }, + { + "epoch": 0.7306273062730627, + "grad_norm": 4.384919166564941, + "learning_rate": 3.049983838570847e-07, + "loss": 0.2679, + "step": 594 + }, + { + "epoch": 0.7318573185731857, + "grad_norm": 3.402482271194458, + "learning_rate": 2.6823105153145125e-07, + "loss": 0.153, + "step": 595 + }, + { + "epoch": 0.7330873308733087, + "grad_norm": 3.2714133262634277, + "learning_rate": 2.3379444289913455e-07, + "loss": 0.1523, + "step": 596 + }, + { + "epoch": 0.7343173431734318, + "grad_norm": 3.7031033039093018, + "learning_rate": 2.016968054041546e-07, + "loss": 0.1988, + "step": 597 + }, + { + "epoch": 0.7355473554735548, + "grad_norm": 3.205819606781006, + "learning_rate": 1.7194582631535505e-07, + "loss": 0.157, + "step": 598 + }, + { + "epoch": 0.7367773677736777, + "grad_norm": 3.5272719860076904, + "learning_rate": 1.4454863088532388e-07, + "loss": 0.146, + "step": 599 + }, + { + "epoch": 0.7380073800738007, + "grad_norm": 3.9655210971832275, + "learning_rate": 1.19511780643915e-07, + "loss": 0.2043, + "step": 600 + }, + { + "epoch": 0.7392373923739237, + "grad_norm": 3.6133296489715576, + "learning_rate": 9.684127182679637e-08, + "loss": 0.1508, + "step": 601 + }, + { + "epoch": 0.7404674046740467, + "grad_norm": 3.886355400085449, + "learning_rate": 7.65425339393644e-08, + "loss": 0.1946, + "step": 602 + }, + { + "epoch": 0.7416974169741697, + "grad_norm": 3.247494697570801, + "learning_rate": 5.862042845640403e-08, + "loss": 0.1481, + "step": 603 + }, + { + "epoch": 0.7429274292742928, + "grad_norm": 2.8707709312438965, + "learning_rate": 4.3079247657777935e-08, + "loss": 0.0995, + "step": 604 + }, + { + "epoch": 0.7441574415744158, + "grad_norm": 3.4677045345306396, + "learning_rate": 2.992271360043986e-08, + "loss": 0.149, + "step": 605 + }, + { + "epoch": 0.7453874538745388, + "grad_norm": 3.709595203399658, + "learning_rate": 1.9153977227022168e-08, + "loss": 0.1686, + "step": 606 + }, + { + "epoch": 0.7466174661746617, + "grad_norm": 3.375216245651245, + "learning_rate": 1.0775617611189504e-08, + "loss": 0.1555, + "step": 607 + }, + { + "epoch": 0.7478474784747847, + "grad_norm": 3.6172735691070557, + "learning_rate": 4.7896413399639575e-09, + "loss": 0.1662, + "step": 608 + }, + { + "epoch": 0.7490774907749077, + "grad_norm": 3.2114450931549072, + "learning_rate": 1.1974820331517313e-09, + "loss": 0.1217, + "step": 609 + }, + { + "epoch": 0.7490774907749077, + "eval_loss": 0.09172860532999039, + "eval_runtime": 54.6078, + "eval_samples_per_second": 27.578, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8783002649533855, + "eval_sts-test_pearson_dot": 0.8649442898053734, + "eval_sts-test_pearson_euclidean": 0.9087900719100788, + "eval_sts-test_pearson_manhattan": 0.9094421155597223, + "eval_sts-test_pearson_max": 0.9094421155597223, + "eval_sts-test_spearman_cosine": 0.906385783044811, + "eval_sts-test_spearman_dot": 0.8744832353165417, + "eval_sts-test_spearman_euclidean": 0.9064414983652431, + "eval_sts-test_spearman_manhattan": 0.9068289547863699, + "eval_sts-test_spearman_max": 0.9068289547863699, + "step": 609 + }, + { + "epoch": 0.7503075030750308, + "grad_norm": 2.619407892227173, + "learning_rate": 2e-05, + "loss": 0.0748, + "step": 610 + }, + { + "epoch": 0.7515375153751538, + "grad_norm": 3.702465057373047, + "learning_rate": 1.9998802517966852e-05, + "loss": 0.1723, + "step": 611 + }, + { + "epoch": 0.7527675276752768, + "grad_norm": 4.303463459014893, + "learning_rate": 1.9995210358660037e-05, + "loss": 0.2354, + "step": 612 + }, + { + "epoch": 0.7539975399753998, + "grad_norm": 3.5168609619140625, + "learning_rate": 1.9989224382388813e-05, + "loss": 0.1315, + "step": 613 + }, + { + "epoch": 0.7552275522755227, + "grad_norm": 4.486151218414307, + "learning_rate": 1.9980846022772978e-05, + "loss": 0.2913, + "step": 614 + }, + { + "epoch": 0.7564575645756457, + "grad_norm": 2.784924030303955, + "learning_rate": 1.9970077286399564e-05, + "loss": 0.0991, + "step": 615 + }, + { + "epoch": 0.7576875768757687, + "grad_norm": 2.8905365467071533, + "learning_rate": 1.9956920752342226e-05, + "loss": 0.1052, + "step": 616 + }, + { + "epoch": 0.7589175891758918, + "grad_norm": 3.176091432571411, + "learning_rate": 1.9941379571543597e-05, + "loss": 0.1496, + "step": 617 + }, + { + "epoch": 0.7601476014760148, + "grad_norm": 3.715683937072754, + "learning_rate": 1.9923457466060637e-05, + "loss": 0.1399, + "step": 618 + }, + { + "epoch": 0.7613776137761378, + "grad_norm": 3.500291347503662, + "learning_rate": 1.9903158728173206e-05, + "loss": 0.1329, + "step": 619 + }, + { + "epoch": 0.7626076260762608, + "grad_norm": 3.6510071754455566, + "learning_rate": 1.9880488219356086e-05, + "loss": 0.2287, + "step": 620 + }, + { + "epoch": 0.7638376383763837, + "grad_norm": 2.6023175716400146, + "learning_rate": 1.9855451369114677e-05, + "loss": 0.1085, + "step": 621 + }, + { + "epoch": 0.7650676506765067, + "grad_norm": 3.7490828037261963, + "learning_rate": 1.9828054173684643e-05, + "loss": 0.1864, + "step": 622 + }, + { + "epoch": 0.7662976629766297, + "grad_norm": 3.3657188415527344, + "learning_rate": 1.9798303194595846e-05, + "loss": 0.1577, + "step": 623 + }, + { + "epoch": 0.7675276752767528, + "grad_norm": 3.729870319366455, + "learning_rate": 1.9766205557100865e-05, + "loss": 0.143, + "step": 624 + }, + { + "epoch": 0.7687576875768758, + "grad_norm": 4.043009281158447, + "learning_rate": 1.9731768948468547e-05, + "loss": 0.1886, + "step": 625 + }, + { + "epoch": 0.7699876998769988, + "grad_norm": 3.5712006092071533, + "learning_rate": 1.9695001616142916e-05, + "loss": 0.1683, + "step": 626 + }, + { + "epoch": 0.7712177121771218, + "grad_norm": 3.5800132751464844, + "learning_rate": 1.965591236576794e-05, + "loss": 0.212, + "step": 627 + }, + { + "epoch": 0.7724477244772447, + "grad_norm": 3.6095833778381348, + "learning_rate": 1.9614510559078622e-05, + "loss": 0.1643, + "step": 628 + }, + { + "epoch": 0.7736777367773677, + "grad_norm": 4.101685523986816, + "learning_rate": 1.95708061116589e-05, + "loss": 0.1632, + "step": 629 + }, + { + "epoch": 0.7749077490774908, + "grad_norm": 3.475429058074951, + "learning_rate": 1.9524809490566878e-05, + "loss": 0.1384, + "step": 630 + }, + { + "epoch": 0.7749077490774908, + "eval_loss": 0.09254728257656097, + "eval_runtime": 54.6557, + "eval_samples_per_second": 27.554, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8771934703209905, + "eval_sts-test_pearson_dot": 0.8615143886353116, + "eval_sts-test_pearson_euclidean": 0.9073193006463275, + "eval_sts-test_pearson_manhattan": 0.908233602851233, + "eval_sts-test_pearson_max": 0.908233602851233, + "eval_sts-test_spearman_cosine": 0.9054128011196391, + "eval_sts-test_spearman_dot": 0.870441346138932, + "eval_sts-test_spearman_euclidean": 0.9052194309191668, + "eval_sts-test_spearman_manhattan": 0.9064029675291854, + "eval_sts-test_spearman_max": 0.9064029675291854, + "step": 630 + }, + { + "epoch": 0.7761377613776138, + "grad_norm": 3.8725736141204834, + "learning_rate": 1.9476531711828027e-05, + "loss": 0.2133, + "step": 631 + }, + { + "epoch": 0.7773677736777368, + "grad_norm": 3.7978456020355225, + "learning_rate": 1.942598433779687e-05, + "loss": 0.1732, + "step": 632 + }, + { + "epoch": 0.7785977859778598, + "grad_norm": 3.1487979888916016, + "learning_rate": 1.9373179474387858e-05, + "loss": 0.1218, + "step": 633 + }, + { + "epoch": 0.7798277982779828, + "grad_norm": 3.349752902984619, + "learning_rate": 1.9318129768176036e-05, + "loss": 0.1581, + "step": 634 + }, + { + "epoch": 0.7810578105781057, + "grad_norm": 2.9575674533843994, + "learning_rate": 1.926084840336821e-05, + "loss": 0.1337, + "step": 635 + }, + { + "epoch": 0.7822878228782287, + "grad_norm": 3.5924367904663086, + "learning_rate": 1.9201349098645433e-05, + "loss": 0.1859, + "step": 636 + }, + { + "epoch": 0.7835178351783518, + "grad_norm": 3.343787670135498, + "learning_rate": 1.9139646103877378e-05, + "loss": 0.1616, + "step": 637 + }, + { + "epoch": 0.7847478474784748, + "grad_norm": 3.2404732704162598, + "learning_rate": 1.9075754196709567e-05, + "loss": 0.1799, + "step": 638 + }, + { + "epoch": 0.7859778597785978, + "grad_norm": 2.723644256591797, + "learning_rate": 1.9009688679024195e-05, + "loss": 0.1193, + "step": 639 + }, + { + "epoch": 0.7872078720787208, + "grad_norm": 3.4389986991882324, + "learning_rate": 1.894146537327533e-05, + "loss": 0.1471, + "step": 640 + }, + { + "epoch": 0.7884378843788438, + "grad_norm": 2.983527898788452, + "learning_rate": 1.8871100618699553e-05, + "loss": 0.1235, + "step": 641 + }, + { + "epoch": 0.7896678966789668, + "grad_norm": 3.521028757095337, + "learning_rate": 1.8798611267402748e-05, + "loss": 0.1221, + "step": 642 + }, + { + "epoch": 0.7908979089790897, + "grad_norm": 3.430474042892456, + "learning_rate": 1.872401468032406e-05, + "loss": 0.1379, + "step": 643 + }, + { + "epoch": 0.7921279212792128, + "grad_norm": 4.5535783767700195, + "learning_rate": 1.864732872307804e-05, + "loss": 0.238, + "step": 644 + }, + { + "epoch": 0.7933579335793358, + "grad_norm": 3.735515594482422, + "learning_rate": 1.8568571761675893e-05, + "loss": 0.1671, + "step": 645 + }, + { + "epoch": 0.7945879458794588, + "grad_norm": 3.2260425090789795, + "learning_rate": 1.8487762658126865e-05, + "loss": 0.1652, + "step": 646 + }, + { + "epoch": 0.7958179581795818, + "grad_norm": 3.5460124015808105, + "learning_rate": 1.8404920765920898e-05, + "loss": 0.1828, + "step": 647 + }, + { + "epoch": 0.7970479704797048, + "grad_norm": 3.820298671722412, + "learning_rate": 1.8320065925393468e-05, + "loss": 0.2207, + "step": 648 + }, + { + "epoch": 0.7982779827798278, + "grad_norm": 4.267491340637207, + "learning_rate": 1.8233218458973984e-05, + "loss": 0.2109, + "step": 649 + }, + { + "epoch": 0.7995079950799509, + "grad_norm": 2.6194908618927, + "learning_rate": 1.8144399166318578e-05, + "loss": 0.1105, + "step": 650 + }, + { + "epoch": 0.8007380073800738, + "grad_norm": 3.2872302532196045, + "learning_rate": 1.8053629319328662e-05, + "loss": 0.129, + "step": 651 + }, + { + "epoch": 0.8007380073800738, + "eval_loss": 0.09332668036222458, + "eval_runtime": 54.6488, + "eval_samples_per_second": 27.558, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8751186892033792, + "eval_sts-test_pearson_dot": 0.8578337888495595, + "eval_sts-test_pearson_euclidean": 0.9068117864702949, + "eval_sts-test_pearson_manhattan": 0.9075917449202227, + "eval_sts-test_pearson_max": 0.9075917449202227, + "eval_sts-test_spearman_cosine": 0.9069310771648651, + "eval_sts-test_spearman_dot": 0.8672049349151059, + "eval_sts-test_spearman_euclidean": 0.9061272550077528, + "eval_sts-test_spearman_manhattan": 0.9067222230279515, + "eval_sts-test_spearman_max": 0.9069310771648651, + "step": 651 + }, + { + "epoch": 0.8019680196801968, + "grad_norm": 3.504014492034912, + "learning_rate": 1.796093065705644e-05, + "loss": 0.1633, + "step": 652 + }, + { + "epoch": 0.8031980319803198, + "grad_norm": 3.93658447265625, + "learning_rate": 1.786632538049842e-05, + "loss": 0.201, + "step": 653 + }, + { + "epoch": 0.8044280442804428, + "grad_norm": 2.9908487796783447, + "learning_rate": 1.7769836147278378e-05, + "loss": 0.1041, + "step": 654 + }, + { + "epoch": 0.8056580565805658, + "grad_norm": 3.6725528240203857, + "learning_rate": 1.7671486066220968e-05, + "loss": 0.1838, + "step": 655 + }, + { + "epoch": 0.8068880688806888, + "grad_norm": 4.807125568389893, + "learning_rate": 1.757129869181718e-05, + "loss": 0.3044, + "step": 656 + }, + { + "epoch": 0.8081180811808119, + "grad_norm": 3.946976900100708, + "learning_rate": 1.746929801858317e-05, + "loss": 0.1736, + "step": 657 + }, + { + "epoch": 0.8093480934809348, + "grad_norm": 3.81947660446167, + "learning_rate": 1.7365508475313667e-05, + "loss": 0.1909, + "step": 658 + }, + { + "epoch": 0.8105781057810578, + "grad_norm": 3.4830641746520996, + "learning_rate": 1.7259954919231313e-05, + "loss": 0.1413, + "step": 659 + }, + { + "epoch": 0.8118081180811808, + "grad_norm": 2.9245359897613525, + "learning_rate": 1.7152662630033506e-05, + "loss": 0.1138, + "step": 660 + }, + { + "epoch": 0.8130381303813038, + "grad_norm": 3.203655481338501, + "learning_rate": 1.7043657303837965e-05, + "loss": 0.1163, + "step": 661 + }, + { + "epoch": 0.8142681426814268, + "grad_norm": 3.563831329345703, + "learning_rate": 1.693296504702862e-05, + "loss": 0.1725, + "step": 662 + }, + { + "epoch": 0.8154981549815498, + "grad_norm": 4.15773344039917, + "learning_rate": 1.6820612370003225e-05, + "loss": 0.2248, + "step": 663 + }, + { + "epoch": 0.8167281672816729, + "grad_norm": 3.2956302165985107, + "learning_rate": 1.670662618082419e-05, + "loss": 0.1019, + "step": 664 + }, + { + "epoch": 0.8179581795817958, + "grad_norm": 3.1350133419036865, + "learning_rate": 1.659103377877423e-05, + "loss": 0.1138, + "step": 665 + }, + { + "epoch": 0.8191881918819188, + "grad_norm": 3.6395132541656494, + "learning_rate": 1.6473862847818283e-05, + "loss": 0.1652, + "step": 666 + }, + { + "epoch": 0.8204182041820418, + "grad_norm": 2.9447243213653564, + "learning_rate": 1.635514144997326e-05, + "loss": 0.1361, + "step": 667 + }, + { + "epoch": 0.8216482164821648, + "grad_norm": 3.6803388595581055, + "learning_rate": 1.623489801858734e-05, + "loss": 0.1769, + "step": 668 + }, + { + "epoch": 0.8228782287822878, + "grad_norm": 3.4124131202697754, + "learning_rate": 1.611316135153026e-05, + "loss": 0.1241, + "step": 669 + }, + { + "epoch": 0.8241082410824109, + "grad_norm": 3.506819725036621, + "learning_rate": 1.5989960604296335e-05, + "loss": 0.1683, + "step": 670 + }, + { + "epoch": 0.8253382533825339, + "grad_norm": 3.3710439205169678, + "learning_rate": 1.586532528302184e-05, + "loss": 0.1315, + "step": 671 + }, + { + "epoch": 0.8265682656826568, + "grad_norm": 2.8190174102783203, + "learning_rate": 1.5739285237418327e-05, + "loss": 0.1046, + "step": 672 + }, + { + "epoch": 0.8265682656826568, + "eval_loss": 0.0940202921628952, + "eval_runtime": 54.5539, + "eval_samples_per_second": 27.606, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8773521416381629, + "eval_sts-test_pearson_dot": 0.8651518375936682, + "eval_sts-test_pearson_euclidean": 0.9076325492477353, + "eval_sts-test_pearson_manhattan": 0.9074800935961884, + "eval_sts-test_pearson_max": 0.9076325492477353, + "eval_sts-test_spearman_cosine": 0.9054711567644934, + "eval_sts-test_spearman_dot": 0.8731833454269057, + "eval_sts-test_spearman_euclidean": 0.9045861558192172, + "eval_sts-test_spearman_manhattan": 0.9044766047313311, + "eval_sts-test_spearman_max": 0.9054711567644934, + "step": 672 + }, + { + "epoch": 0.8277982779827798, + "grad_norm": 3.8072876930236816, + "learning_rate": 1.5611870653623826e-05, + "loss": 0.1984, + "step": 673 + }, + { + "epoch": 0.8290282902829028, + "grad_norm": 3.769066572189331, + "learning_rate": 1.548311204697332e-05, + "loss": 0.1766, + "step": 674 + }, + { + "epoch": 0.8302583025830258, + "grad_norm": 2.888267993927002, + "learning_rate": 1.535304025469039e-05, + "loss": 0.1245, + "step": 675 + }, + { + "epoch": 0.8314883148831488, + "grad_norm": 3.8960931301116943, + "learning_rate": 1.5221686428501925e-05, + "loss": 0.1953, + "step": 676 + }, + { + "epoch": 0.8327183271832719, + "grad_norm": 3.544792413711548, + "learning_rate": 1.5089082027177293e-05, + "loss": 0.1506, + "step": 677 + }, + { + "epoch": 0.8339483394833949, + "grad_norm": 3.2209813594818115, + "learning_rate": 1.4955258808994093e-05, + "loss": 0.1145, + "step": 678 + }, + { + "epoch": 0.8351783517835178, + "grad_norm": 3.3462533950805664, + "learning_rate": 1.4820248824132218e-05, + "loss": 0.1366, + "step": 679 + }, + { + "epoch": 0.8364083640836408, + "grad_norm": 3.4571852684020996, + "learning_rate": 1.4684084406997898e-05, + "loss": 0.1071, + "step": 680 + }, + { + "epoch": 0.8376383763837638, + "grad_norm": 4.009058952331543, + "learning_rate": 1.4546798168479746e-05, + "loss": 0.2142, + "step": 681 + }, + { + "epoch": 0.8388683886838868, + "grad_norm": 3.7418360710144043, + "learning_rate": 1.4408422988138585e-05, + "loss": 0.2029, + "step": 682 + }, + { + "epoch": 0.8400984009840098, + "grad_norm": 3.1473541259765625, + "learning_rate": 1.4268992006332845e-05, + "loss": 0.1171, + "step": 683 + }, + { + "epoch": 0.8413284132841329, + "grad_norm": 3.88065242767334, + "learning_rate": 1.4128538616281653e-05, + "loss": 0.176, + "step": 684 + }, + { + "epoch": 0.8425584255842559, + "grad_norm": 2.8190746307373047, + "learning_rate": 1.398709645606724e-05, + "loss": 0.1052, + "step": 685 + }, + { + "epoch": 0.8437884378843789, + "grad_norm": 3.612539052963257, + "learning_rate": 1.3844699400578696e-05, + "loss": 0.1892, + "step": 686 + }, + { + "epoch": 0.8450184501845018, + "grad_norm": 3.7492446899414062, + "learning_rate": 1.3701381553399142e-05, + "loss": 0.1499, + "step": 687 + }, + { + "epoch": 0.8462484624846248, + "grad_norm": 3.3772854804992676, + "learning_rate": 1.3557177238637982e-05, + "loss": 0.1414, + "step": 688 + }, + { + "epoch": 0.8474784747847478, + "grad_norm": 3.1224844455718994, + "learning_rate": 1.3412120992710418e-05, + "loss": 0.1193, + "step": 689 + }, + { + "epoch": 0.8487084870848709, + "grad_norm": 3.244182586669922, + "learning_rate": 1.3266247556066122e-05, + "loss": 0.1516, + "step": 690 + }, + { + "epoch": 0.8499384993849939, + "grad_norm": 3.860499382019043, + "learning_rate": 1.3119591864868977e-05, + "loss": 0.1552, + "step": 691 + }, + { + "epoch": 0.8511685116851169, + "grad_norm": 2.9811863899230957, + "learning_rate": 1.2972189042630039e-05, + "loss": 0.1168, + "step": 692 + }, + { + "epoch": 0.8523985239852399, + "grad_norm": 3.8969922065734863, + "learning_rate": 1.2824074391795578e-05, + "loss": 0.2326, + "step": 693 + }, + { + "epoch": 0.8523985239852399, + "eval_loss": 0.09323134273290634, + "eval_runtime": 54.5627, + "eval_samples_per_second": 27.601, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8788852740353275, + "eval_sts-test_pearson_dot": 0.864663524772969, + "eval_sts-test_pearson_euclidean": 0.9093934738006354, + "eval_sts-test_pearson_manhattan": 0.9091635226202384, + "eval_sts-test_pearson_max": 0.9093934738006354, + "eval_sts-test_spearman_cosine": 0.9071493738179324, + "eval_sts-test_spearman_dot": 0.8718393755447993, + "eval_sts-test_spearman_euclidean": 0.9067696593650264, + "eval_sts-test_spearman_manhattan": 0.9064368442340587, + "eval_sts-test_spearman_max": 0.9071493738179324, + "step": 693 + }, + { + "epoch": 0.8536285362853628, + "grad_norm": 3.883828639984131, + "learning_rate": 1.2675283385292212e-05, + "loss": 0.2112, + "step": 694 + }, + { + "epoch": 0.8548585485854858, + "grad_norm": 2.648932695388794, + "learning_rate": 1.2525851658031349e-05, + "loss": 0.0835, + "step": 695 + }, + { + "epoch": 0.8560885608856088, + "grad_norm": 3.8667919635772705, + "learning_rate": 1.2375814998374708e-05, + "loss": 0.1512, + "step": 696 + }, + { + "epoch": 0.8573185731857319, + "grad_norm": 3.028184175491333, + "learning_rate": 1.2225209339563138e-05, + "loss": 0.1379, + "step": 697 + }, + { + "epoch": 0.8585485854858549, + "grad_norm": 3.003844738006592, + "learning_rate": 1.2074070751110753e-05, + "loss": 0.1045, + "step": 698 + }, + { + "epoch": 0.8597785977859779, + "grad_norm": 4.526356220245361, + "learning_rate": 1.192243543016637e-05, + "loss": 0.2045, + "step": 699 + }, + { + "epoch": 0.8610086100861009, + "grad_norm": 3.4816009998321533, + "learning_rate": 1.1770339692844482e-05, + "loss": 0.1909, + "step": 700 + }, + { + "epoch": 0.8622386223862238, + "grad_norm": 3.9605863094329834, + "learning_rate": 1.1617819965527657e-05, + "loss": 0.1895, + "step": 701 + }, + { + "epoch": 0.8634686346863468, + "grad_norm": 3.9873011112213135, + "learning_rate": 1.1464912776142495e-05, + "loss": 0.2077, + "step": 702 + }, + { + "epoch": 0.8646986469864698, + "grad_norm": 3.3017494678497314, + "learning_rate": 1.1311654745411424e-05, + "loss": 0.1199, + "step": 703 + }, + { + "epoch": 0.8659286592865929, + "grad_norm": 3.6814589500427246, + "learning_rate": 1.1158082578082089e-05, + "loss": 0.1606, + "step": 704 + }, + { + "epoch": 0.8671586715867159, + "grad_norm": 3.539412260055542, + "learning_rate": 1.1004233054136717e-05, + "loss": 0.1501, + "step": 705 + }, + { + "epoch": 0.8683886838868389, + "grad_norm": 3.4747865200042725, + "learning_rate": 1.0850143019983476e-05, + "loss": 0.1711, + "step": 706 + }, + { + "epoch": 0.8696186961869619, + "grad_norm": 3.6860110759735107, + "learning_rate": 1.0695849379631816e-05, + "loss": 0.222, + "step": 707 + }, + { + "epoch": 0.8708487084870848, + "grad_norm": 3.238083600997925, + "learning_rate": 1.0541389085854174e-05, + "loss": 0.1414, + "step": 708 + }, + { + "epoch": 0.8720787207872078, + "grad_norm": 4.44494104385376, + "learning_rate": 1.0386799131335898e-05, + "loss": 0.1972, + "step": 709 + }, + { + "epoch": 0.8733087330873309, + "grad_norm": 3.207988739013672, + "learning_rate": 1.0232116539815561e-05, + "loss": 0.1074, + "step": 710 + }, + { + "epoch": 0.8745387453874539, + "grad_norm": 3.7489352226257324, + "learning_rate": 1.0077378357218023e-05, + "loss": 0.2044, + "step": 711 + }, + { + "epoch": 0.8757687576875769, + "grad_norm": 2.951228141784668, + "learning_rate": 9.92262164278198e-06, + "loss": 0.0997, + "step": 712 + }, + { + "epoch": 0.8769987699876999, + "grad_norm": 3.126646041870117, + "learning_rate": 9.76788346018444e-06, + "loss": 0.1178, + "step": 713 + }, + { + "epoch": 0.8782287822878229, + "grad_norm": 3.140065908432007, + "learning_rate": 9.61320086866412e-06, + "loss": 0.1376, + "step": 714 + }, + { + "epoch": 0.8782287822878229, + "eval_loss": 0.09289773553609848, + "eval_runtime": 54.5909, + "eval_samples_per_second": 27.587, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8778257617860821, + "eval_sts-test_pearson_dot": 0.8615556182815335, + "eval_sts-test_pearson_euclidean": 0.90830154034185, + "eval_sts-test_pearson_manhattan": 0.9082439956690477, + "eval_sts-test_pearson_max": 0.90830154034185, + "eval_sts-test_spearman_cosine": 0.9058249602370539, + "eval_sts-test_spearman_dot": 0.8679172854939341, + "eval_sts-test_spearman_euclidean": 0.9054231365232531, + "eval_sts-test_spearman_manhattan": 0.9050758241213619, + "eval_sts-test_spearman_max": 0.9058249602370539, + "step": 714 + }, + { + "epoch": 0.8794587945879458, + "grad_norm": 3.2434728145599365, + "learning_rate": 9.458610914145828e-06, + "loss": 0.1302, + "step": 715 + }, + { + "epoch": 0.8806888068880688, + "grad_norm": 2.9910123348236084, + "learning_rate": 9.304150620368189e-06, + "loss": 0.1252, + "step": 716 + }, + { + "epoch": 0.8819188191881919, + "grad_norm": 3.8356688022613525, + "learning_rate": 9.14985698001654e-06, + "loss": 0.2365, + "step": 717 + }, + { + "epoch": 0.8831488314883149, + "grad_norm": 3.2038049697875977, + "learning_rate": 8.995766945863286e-06, + "loss": 0.1405, + "step": 718 + }, + { + "epoch": 0.8843788437884379, + "grad_norm": 3.7703609466552734, + "learning_rate": 8.841917421917915e-06, + "loss": 0.1806, + "step": 719 + }, + { + "epoch": 0.8856088560885609, + "grad_norm": 3.7256062030792236, + "learning_rate": 8.688345254588579e-06, + "loss": 0.1495, + "step": 720 + }, + { + "epoch": 0.8868388683886839, + "grad_norm": 4.105401039123535, + "learning_rate": 8.535087223857506e-06, + "loss": 0.1987, + "step": 721 + }, + { + "epoch": 0.8880688806888068, + "grad_norm": 2.8106296062469482, + "learning_rate": 8.382180034472362e-06, + "loss": 0.096, + "step": 722 + }, + { + "epoch": 0.8892988929889298, + "grad_norm": 3.8372719287872314, + "learning_rate": 8.22966030715552e-06, + "loss": 0.1728, + "step": 723 + }, + { + "epoch": 0.8905289052890529, + "grad_norm": 3.965484857559204, + "learning_rate": 8.077564569833633e-06, + "loss": 0.2104, + "step": 724 + }, + { + "epoch": 0.8917589175891759, + "grad_norm": 3.417832136154175, + "learning_rate": 7.925929248889262e-06, + "loss": 0.1562, + "step": 725 + }, + { + "epoch": 0.8929889298892989, + "grad_norm": 3.7467880249023438, + "learning_rate": 7.774790660436866e-06, + "loss": 0.1358, + "step": 726 + }, + { + "epoch": 0.8942189421894219, + "grad_norm": 3.463932752609253, + "learning_rate": 7.624185001625294e-06, + "loss": 0.1723, + "step": 727 + }, + { + "epoch": 0.8954489544895449, + "grad_norm": 3.8118107318878174, + "learning_rate": 7.4741483419686525e-06, + "loss": 0.1947, + "step": 728 + }, + { + "epoch": 0.8966789667896679, + "grad_norm": 3.5355310440063477, + "learning_rate": 7.324716614707792e-06, + "loss": 0.1572, + "step": 729 + }, + { + "epoch": 0.897908979089791, + "grad_norm": 2.99922513961792, + "learning_rate": 7.175925608204424e-06, + "loss": 0.1124, + "step": 730 + }, + { + "epoch": 0.8991389913899139, + "grad_norm": 4.25166130065918, + "learning_rate": 7.0278109573699515e-06, + "loss": 0.2272, + "step": 731 + }, + { + "epoch": 0.9003690036900369, + "grad_norm": 3.3319239616394043, + "learning_rate": 6.880408135131014e-06, + "loss": 0.1356, + "step": 732 + }, + { + "epoch": 0.9015990159901599, + "grad_norm": 3.799717664718628, + "learning_rate": 6.733752443933879e-06, + "loss": 0.1816, + "step": 733 + }, + { + "epoch": 0.9028290282902829, + "grad_norm": 2.819953203201294, + "learning_rate": 6.587879007289572e-06, + "loss": 0.1011, + "step": 734 + }, + { + "epoch": 0.9040590405904059, + "grad_norm": 3.7279915809631348, + "learning_rate": 6.442822761362009e-06, + "loss": 0.124, + "step": 735 + }, + { + "epoch": 0.9040590405904059, + "eval_loss": 0.09114937484264374, + "eval_runtime": 54.5694, + "eval_samples_per_second": 27.598, + "eval_steps_per_second": 0.22, + "eval_sts-test_pearson_cosine": 0.8753500420421221, + "eval_sts-test_pearson_dot": 0.8591038680239624, + "eval_sts-test_pearson_euclidean": 0.9065823860587949, + "eval_sts-test_pearson_manhattan": 0.9066636021783758, + "eval_sts-test_pearson_max": 0.9066636021783758, + "eval_sts-test_spearman_cosine": 0.9050749738473955, + "eval_sts-test_spearman_dot": 0.8668577546292313, + "eval_sts-test_spearman_euclidean": 0.9046445114640715, + "eval_sts-test_spearman_manhattan": 0.903748501708493, + "eval_sts-test_spearman_max": 0.9050749738473955, + "step": 735 + }, + { + "epoch": 0.9052890528905289, + "grad_norm": 4.3227925300598145, + "learning_rate": 6.29861844660086e-06, + "loss": 0.1873, + "step": 736 + }, + { + "epoch": 0.906519065190652, + "grad_norm": 2.7029824256896973, + "learning_rate": 6.155300599421305e-06, + "loss": 0.0702, + "step": 737 + }, + { + "epoch": 0.9077490774907749, + "grad_norm": 3.703040599822998, + "learning_rate": 6.012903543932762e-06, + "loss": 0.15, + "step": 738 + } + ], + "logging_steps": 1, + "max_steps": 813, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 82, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 320, + "trial_name": null, + "trial_params": null +}