| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.899159663865546, | |
| "eval_steps": 8, | |
| "global_step": 116, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03361344537815126, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0282, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03361344537815126, | |
| "eval_loss": 3.090120792388916, | |
| "eval_runtime": 9.9371, | |
| "eval_samples_per_second": 5.032, | |
| "eval_steps_per_second": 0.704, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06722689075630252, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 4e-05, | |
| "loss": 3.0737, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10084033613445378, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 6e-05, | |
| "loss": 3.292, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13445378151260504, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 8e-05, | |
| "loss": 3.0494, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2068, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20168067226890757, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.00012, | |
| "loss": 3.2567, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00014, | |
| "loss": 2.9936, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.00016, | |
| "loss": 3.0212, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "eval_loss": 2.9598207473754883, | |
| "eval_runtime": 10.5329, | |
| "eval_samples_per_second": 4.747, | |
| "eval_steps_per_second": 0.665, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3025210084033613, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00018, | |
| "loss": 2.9687, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0002, | |
| "loss": 3.0585, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3697478991596639, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00019995608365087946, | |
| "loss": 2.8995, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.40336134453781514, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00019982437317643217, | |
| "loss": 2.7126, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4369747899159664, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0001996049842615217, | |
| "loss": 2.67, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.00019929810960135172, | |
| "loss": 2.5902, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.0001989040187322164, | |
| "loss": 2.6126, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.00019842305779475968, | |
| "loss": 2.6598, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "eval_loss": 2.589160680770874, | |
| "eval_runtime": 10.6131, | |
| "eval_samples_per_second": 4.711, | |
| "eval_steps_per_second": 0.66, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.0001978556492299504, | |
| "loss": 2.5809, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6050420168067226, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.0001972022914080411, | |
| "loss": 2.6425, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6386554621848739, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.00019646355819083589, | |
| "loss": 2.5424, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00019564009842765225, | |
| "loss": 2.4965, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.00019473263538541914, | |
| "loss": 2.4281, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7394957983193278, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.0001937419661134121, | |
| "loss": 2.412, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.773109243697479, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.00019266896074318334, | |
| "loss": 2.5341, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00019151456172430183, | |
| "loss": 2.155, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "eval_loss": 2.261070489883423, | |
| "eval_runtime": 10.494, | |
| "eval_samples_per_second": 4.765, | |
| "eval_steps_per_second": 0.667, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00019027978299657436, | |
| "loss": 2.1934, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8739495798319328, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.00018896570909947475, | |
| "loss": 2.3187, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.907563025210084, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.0001875734942195637, | |
| "loss": 2.4368, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.00018610436117673555, | |
| "loss": 2.218, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9747899159663865, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.0001845596003501826, | |
| "loss": 2.4088, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0084033613445378, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.0001829405685450202, | |
| "loss": 2.3832, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0420168067226891, | |
| "grad_norm": 0.2109375, | |
| "learning_rate": 0.00018124868780056814, | |
| "loss": 2.2727, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00017948544414133534, | |
| "loss": 2.262, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "eval_loss": 2.202693462371826, | |
| "eval_runtime": 10.5469, | |
| "eval_samples_per_second": 4.741, | |
| "eval_steps_per_second": 0.664, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.1092436974789917, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 0.00017765238627180424, | |
| "loss": 2.2559, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 0.00017575112421616202, | |
| "loss": 2.1156, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.00017378332790417273, | |
| "loss": 2.3003, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2100840336134453, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.00017175072570443312, | |
| "loss": 2.2861, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.2436974789915967, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 0.00016965510290629972, | |
| "loss": 2.3813, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2773109243697478, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 0.00016749830015182107, | |
| "loss": 2.2533, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.3109243697478992, | |
| "grad_norm": 0.1884765625, | |
| "learning_rate": 0.00016528221181905217, | |
| "loss": 2.3311, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 0.00016300878435817113, | |
| "loss": 2.1765, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "eval_loss": 2.1833150386810303, | |
| "eval_runtime": 10.5583, | |
| "eval_samples_per_second": 4.736, | |
| "eval_steps_per_second": 0.663, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3781512605042017, | |
| "grad_norm": 0.203125, | |
| "learning_rate": 0.00016068001458185936, | |
| "loss": 2.207, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.0001582979479114472, | |
| "loss": 2.0924, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4453781512605042, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 0.00015586467658036524, | |
| "loss": 2.3548, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.4789915966386555, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 0.0001533823377964791, | |
| "loss": 2.052, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.5126050420168067, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 0.00015085311186492206, | |
| "loss": 2.0606, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5462184873949578, | |
| "grad_norm": 0.189453125, | |
| "learning_rate": 0.00014827922027307451, | |
| "loss": 2.124, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.5798319327731094, | |
| "grad_norm": 0.193359375, | |
| "learning_rate": 0.0001456629237393713, | |
| "loss": 2.2124, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 0.00014300652022765207, | |
| "loss": 2.2249, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "eval_loss": 2.174049139022827, | |
| "eval_runtime": 10.5335, | |
| "eval_samples_per_second": 4.747, | |
| "eval_steps_per_second": 0.665, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 0.00014031234292879725, | |
| "loss": 2.2029, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.680672268907563, | |
| "grad_norm": 0.1826171875, | |
| "learning_rate": 0.00013758275821142382, | |
| "loss": 2.0138, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 0.0001348201635434399, | |
| "loss": 2.2726, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.7478991596638656, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 0.00013202698538628376, | |
| "loss": 2.1516, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.7815126050420167, | |
| "grad_norm": 0.1826171875, | |
| "learning_rate": 0.00012920567706369758, | |
| "loss": 2.2919, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.815126050420168, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 0.00012635871660690676, | |
| "loss": 2.2489, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.8487394957983194, | |
| "grad_norm": 0.177734375, | |
| "learning_rate": 0.00012348860457809838, | |
| "loss": 2.2379, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 0.00012059786187410984, | |
| "loss": 2.1377, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "eval_loss": 2.16938853263855, | |
| "eval_runtime": 10.4962, | |
| "eval_samples_per_second": 4.764, | |
| "eval_steps_per_second": 0.667, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9159663865546217, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 0.0001176890275122573, | |
| "loss": 1.9655, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.949579831932773, | |
| "grad_norm": 0.2041015625, | |
| "learning_rate": 0.00011476465640024814, | |
| "loss": 2.2597, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.9831932773109244, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.00011182731709213659, | |
| "loss": 2.0952, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.0168067226890756, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 0.00010887958953229349, | |
| "loss": 2.0645, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0504201680672267, | |
| "grad_norm": 0.1923828125, | |
| "learning_rate": 0.00010592406278937144, | |
| "loss": 2.3231, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.0840336134453783, | |
| "grad_norm": 0.1728515625, | |
| "learning_rate": 0.00010296333278225599, | |
| "loss": 2.2175, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.18359375, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1928, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 9.703666721774402e-05, | |
| "loss": 2.0569, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "eval_loss": 2.166851758956909, | |
| "eval_runtime": 10.5358, | |
| "eval_samples_per_second": 4.746, | |
| "eval_steps_per_second": 0.664, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.184873949579832, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 9.407593721062859e-05, | |
| "loss": 2.1509, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2184873949579833, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 9.112041046770653e-05, | |
| "loss": 2.1184, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.2521008403361344, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 8.817268290786343e-05, | |
| "loss": 2.1504, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 8.523534359975189e-05, | |
| "loss": 2.0963, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.3193277310924367, | |
| "grad_norm": 0.1875, | |
| "learning_rate": 8.231097248774274e-05, | |
| "loss": 2.0506, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.1728515625, | |
| "learning_rate": 7.940213812589018e-05, | |
| "loss": 2.0421, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.3865546218487395, | |
| "grad_norm": 0.1708984375, | |
| "learning_rate": 7.651139542190164e-05, | |
| "loss": 2.0611, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "grad_norm": 0.1806640625, | |
| "learning_rate": 7.364128339309326e-05, | |
| "loss": 2.1184, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "eval_loss": 2.163745164871216, | |
| "eval_runtime": 10.6427, | |
| "eval_samples_per_second": 4.698, | |
| "eval_steps_per_second": 0.658, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.453781512605042, | |
| "grad_norm": 0.2001953125, | |
| "learning_rate": 7.079432293630244e-05, | |
| "loss": 2.1386, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.4873949579831933, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 6.797301461371625e-05, | |
| "loss": 2.0429, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.5210084033613445, | |
| "grad_norm": 0.2138671875, | |
| "learning_rate": 6.517983645656014e-05, | |
| "loss": 2.0244, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5546218487394956, | |
| "grad_norm": 0.1875, | |
| "learning_rate": 6.24172417885762e-05, | |
| "loss": 2.2017, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.1884765625, | |
| "learning_rate": 5.96876570712028e-05, | |
| "loss": 2.1196, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.6218487394957983, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 5.699347977234799e-05, | |
| "loss": 2.0911, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.6554621848739495, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 5.43370762606287e-05, | |
| "loss": 2.2691, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 5.172077972692553e-05, | |
| "loss": 2.1894, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "eval_loss": 2.1625115871429443, | |
| "eval_runtime": 10.5832, | |
| "eval_samples_per_second": 4.724, | |
| "eval_steps_per_second": 0.661, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.722689075630252, | |
| "grad_norm": 0.1748046875, | |
| "learning_rate": 4.914688813507797e-05, | |
| "loss": 2.1408, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.7563025210084033, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 4.661766220352097e-05, | |
| "loss": 2.3066, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.7899159663865545, | |
| "grad_norm": 0.177734375, | |
| "learning_rate": 4.4135323419634766e-05, | |
| "loss": 1.995, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.2001953125, | |
| "learning_rate": 4.170205208855281e-05, | |
| "loss": 2.1791, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 3.931998541814069e-05, | |
| "loss": 2.0588, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.8907563025210083, | |
| "grad_norm": 0.1953125, | |
| "learning_rate": 3.69912156418289e-05, | |
| "loss": 2.1747, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.92436974789916, | |
| "grad_norm": 0.193359375, | |
| "learning_rate": 3.471778818094785e-05, | |
| "loss": 2.2481, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.957983193277311, | |
| "grad_norm": 0.19921875, | |
| "learning_rate": 3.250169984817897e-05, | |
| "loss": 2.2582, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.957983193277311, | |
| "eval_loss": 2.1615426540374756, | |
| "eval_runtime": 10.5786, | |
| "eval_samples_per_second": 4.727, | |
| "eval_steps_per_second": 0.662, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.991596638655462, | |
| "grad_norm": 0.189453125, | |
| "learning_rate": 3.034489709370033e-05, | |
| "loss": 2.0839, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.0252100840336134, | |
| "grad_norm": 0.2060546875, | |
| "learning_rate": 2.8249274295566864e-05, | |
| "loss": 2.2324, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 2.6216672095827266e-05, | |
| "loss": 2.1676, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.092436974789916, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 2.4248875783837987e-05, | |
| "loss": 2.2775, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.1260504201680672, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 2.234761372819577e-05, | |
| "loss": 2.2849, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.1596638655462184, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 2.0514555858664663e-05, | |
| "loss": 2.0748, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.19327731092437, | |
| "grad_norm": 0.189453125, | |
| "learning_rate": 1.875131219943187e-05, | |
| "loss": 2.0704, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.226890756302521, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 1.7059431454979824e-05, | |
| "loss": 2.0791, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.226890756302521, | |
| "eval_loss": 2.1611685752868652, | |
| "eval_runtime": 10.5365, | |
| "eval_samples_per_second": 4.745, | |
| "eval_steps_per_second": 0.664, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.2605042016806722, | |
| "grad_norm": 0.1884765625, | |
| "learning_rate": 1.5440399649817385e-05, | |
| "loss": 2.1193, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.201171875, | |
| "learning_rate": 1.3895638823264446e-05, | |
| "loss": 2.1647, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.327731092436975, | |
| "grad_norm": 0.18359375, | |
| "learning_rate": 1.2426505780436326e-05, | |
| "loss": 2.2459, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.361344537815126, | |
| "grad_norm": 0.1796875, | |
| "learning_rate": 1.103429090052528e-05, | |
| "loss": 2.2252, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.3949579831932772, | |
| "grad_norm": 0.18359375, | |
| "learning_rate": 9.720217003425647e-06, | |
| "loss": 2.0849, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 8.485438275698154e-06, | |
| "loss": 2.059, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.46218487394958, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 7.331039256816663e-06, | |
| "loss": 2.1378, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 3.495798319327731, | |
| "grad_norm": 0.189453125, | |
| "learning_rate": 6.258033886587911e-06, | |
| "loss": 2.2571, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 3.495798319327731, | |
| "eval_loss": 2.1611392498016357, | |
| "eval_runtime": 10.63, | |
| "eval_samples_per_second": 4.704, | |
| "eval_steps_per_second": 0.659, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.1787109375, | |
| "learning_rate": 5.267364614580861e-06, | |
| "loss": 2.1328, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.5630252100840334, | |
| "grad_norm": 0.1875, | |
| "learning_rate": 4.359901572347758e-06, | |
| "loss": 2.1423, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 3.596638655462185, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 3.5364418091641373e-06, | |
| "loss": 2.2438, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 3.630252100840336, | |
| "grad_norm": 0.1865234375, | |
| "learning_rate": 2.7977085919589254e-06, | |
| "loss": 2.0935, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 3.6638655462184873, | |
| "grad_norm": 0.185546875, | |
| "learning_rate": 2.144350770049597e-06, | |
| "loss": 2.0828, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 3.697478991596639, | |
| "grad_norm": 0.1875, | |
| "learning_rate": 1.576942205240317e-06, | |
| "loss": 1.9742, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.73109243697479, | |
| "grad_norm": 0.205078125, | |
| "learning_rate": 1.0959812677835968e-06, | |
| "loss": 1.9411, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.2138671875, | |
| "learning_rate": 7.018903986483083e-07, | |
| "loss": 2.177, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "eval_loss": 2.1611266136169434, | |
| "eval_runtime": 10.4595, | |
| "eval_samples_per_second": 4.78, | |
| "eval_steps_per_second": 0.669, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.7983193277310923, | |
| "grad_norm": 0.1748046875, | |
| "learning_rate": 3.950157384783104e-07, | |
| "loss": 2.1226, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 3.831932773109244, | |
| "grad_norm": 0.1904296875, | |
| "learning_rate": 1.7562682356786487e-07, | |
| "loss": 2.0753, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.865546218487395, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 4.391634912056519e-08, | |
| "loss": 2.0682, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.899159663865546, | |
| "grad_norm": 0.1923828125, | |
| "learning_rate": 0.0, | |
| "loss": 2.0265, | |
| "step": 116 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 29, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7267177095757824.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |