| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.92436974789916, | |
| "eval_steps": 8, | |
| "global_step": 87, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03361344537815126, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0408, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03361344537815126, | |
| "eval_loss": 3.0278432369232178, | |
| "eval_runtime": 18.2975, | |
| "eval_samples_per_second": 2.733, | |
| "eval_steps_per_second": 0.383, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06722689075630252, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 4e-05, | |
| "loss": 3.042, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10084033613445378, | |
| "grad_norm": 1.625, | |
| "learning_rate": 6e-05, | |
| "loss": 3.3331, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13445378151260504, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 8e-05, | |
| "loss": 3.1531, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7659, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20168067226890757, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00012, | |
| "loss": 2.9096, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 0.00014, | |
| "loss": 3.1465, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 0.00016, | |
| "loss": 2.6971, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "eval_loss": 2.588528633117676, | |
| "eval_runtime": 20.8612, | |
| "eval_samples_per_second": 2.397, | |
| "eval_steps_per_second": 0.336, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3025210084033613, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 0.00018, | |
| "loss": 2.3843, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5628, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3697478991596639, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 0.00019995608365087946, | |
| "loss": 2.4701, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.40336134453781514, | |
| "grad_norm": 0.57421875, | |
| "learning_rate": 0.00019982437317643217, | |
| "loss": 2.205, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4369747899159664, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 0.0001996049842615217, | |
| "loss": 2.2017, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.00019929810960135172, | |
| "loss": 2.1823, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 0.0001989040187322164, | |
| "loss": 2.1578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.00019842305779475968, | |
| "loss": 2.0158, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "eval_loss": 2.1604020595550537, | |
| "eval_runtime": 20.8469, | |
| "eval_samples_per_second": 2.398, | |
| "eval_steps_per_second": 0.336, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.0001978556492299504, | |
| "loss": 2.2053, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6050420168067226, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.0001972022914080411, | |
| "loss": 2.0244, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6386554621848739, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 0.00019646355819083589, | |
| "loss": 2.1662, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 0.00019564009842765225, | |
| "loss": 1.9738, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 0.00019473263538541914, | |
| "loss": 1.993, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7394957983193278, | |
| "grad_norm": 0.11962890625, | |
| "learning_rate": 0.0001937419661134121, | |
| "loss": 1.9879, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.773109243697479, | |
| "grad_norm": 0.1611328125, | |
| "learning_rate": 0.00019266896074318334, | |
| "loss": 2.0819, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "grad_norm": 0.1474609375, | |
| "learning_rate": 0.00019151456172430183, | |
| "loss": 1.9578, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "eval_loss": 2.0713210105895996, | |
| "eval_runtime": 20.87, | |
| "eval_samples_per_second": 2.396, | |
| "eval_steps_per_second": 0.335, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.142578125, | |
| "learning_rate": 0.00019027978299657436, | |
| "loss": 1.9516, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8739495798319328, | |
| "grad_norm": 0.1162109375, | |
| "learning_rate": 0.00018896570909947475, | |
| "loss": 2.0161, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.907563025210084, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.0001875734942195637, | |
| "loss": 2.1119, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.11083984375, | |
| "learning_rate": 0.00018610436117673555, | |
| "loss": 1.9466, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9747899159663865, | |
| "grad_norm": 0.1240234375, | |
| "learning_rate": 0.0001845596003501826, | |
| "loss": 2.0912, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0084033613445378, | |
| "grad_norm": 0.10107421875, | |
| "learning_rate": 0.0001829405685450202, | |
| "loss": 2.0932, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0420168067226891, | |
| "grad_norm": 0.09814453125, | |
| "learning_rate": 0.00018124868780056814, | |
| "loss": 2.0034, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "grad_norm": 0.0810546875, | |
| "learning_rate": 0.00017948544414133534, | |
| "loss": 1.9457, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "eval_loss": 2.0372533798217773, | |
| "eval_runtime": 20.998, | |
| "eval_samples_per_second": 2.381, | |
| "eval_steps_per_second": 0.333, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.1092436974789917, | |
| "grad_norm": 0.06689453125, | |
| "learning_rate": 0.00017765238627180424, | |
| "loss": 1.9816, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.09423828125, | |
| "learning_rate": 0.00017575112421616202, | |
| "loss": 1.9487, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.078125, | |
| "learning_rate": 0.00017378332790417273, | |
| "loss": 1.9633, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2100840336134453, | |
| "grad_norm": 0.0927734375, | |
| "learning_rate": 0.00017175072570443312, | |
| "loss": 1.9589, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.2436974789915967, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.00016965510290629972, | |
| "loss": 1.9918, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2773109243697478, | |
| "grad_norm": 0.08203125, | |
| "learning_rate": 0.00016749830015182107, | |
| "loss": 2.0112, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.3109243697478992, | |
| "grad_norm": 0.0712890625, | |
| "learning_rate": 0.00016528221181905217, | |
| "loss": 1.9123, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "grad_norm": 0.08203125, | |
| "learning_rate": 0.00016300878435817113, | |
| "loss": 1.9927, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "eval_loss": 2.016634941101074, | |
| "eval_runtime": 20.893, | |
| "eval_samples_per_second": 2.393, | |
| "eval_steps_per_second": 0.335, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3781512605042017, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.00016068001458185936, | |
| "loss": 1.9231, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.0001582979479114472, | |
| "loss": 1.9237, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4453781512605042, | |
| "grad_norm": 0.08544921875, | |
| "learning_rate": 0.00015586467658036524, | |
| "loss": 1.9926, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.4789915966386555, | |
| "grad_norm": 0.06396484375, | |
| "learning_rate": 0.0001533823377964791, | |
| "loss": 1.8798, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.5126050420168067, | |
| "grad_norm": 0.09814453125, | |
| "learning_rate": 0.00015085311186492206, | |
| "loss": 1.9035, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5462184873949578, | |
| "grad_norm": 0.07421875, | |
| "learning_rate": 0.00014827922027307451, | |
| "loss": 1.9381, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.5798319327731094, | |
| "grad_norm": 0.0888671875, | |
| "learning_rate": 0.0001456629237393713, | |
| "loss": 1.9413, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "grad_norm": 0.07373046875, | |
| "learning_rate": 0.00014300652022765207, | |
| "loss": 1.9848, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "eval_loss": 2.002051830291748, | |
| "eval_runtime": 20.8487, | |
| "eval_samples_per_second": 2.398, | |
| "eval_steps_per_second": 0.336, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.0732421875, | |
| "learning_rate": 0.00014031234292879725, | |
| "loss": 1.9653, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.680672268907563, | |
| "grad_norm": 0.07666015625, | |
| "learning_rate": 0.00013758275821142382, | |
| "loss": 1.8987, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.0849609375, | |
| "learning_rate": 0.0001348201635434399, | |
| "loss": 1.9558, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.7478991596638656, | |
| "grad_norm": 0.14453125, | |
| "learning_rate": 0.00013202698538628376, | |
| "loss": 1.9345, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.7815126050420167, | |
| "grad_norm": 0.07080078125, | |
| "learning_rate": 0.00012920567706369758, | |
| "loss": 2.042, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.815126050420168, | |
| "grad_norm": 0.06982421875, | |
| "learning_rate": 0.00012635871660690676, | |
| "loss": 1.9398, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.8487394957983194, | |
| "grad_norm": 0.06689453125, | |
| "learning_rate": 0.00012348860457809838, | |
| "loss": 1.9625, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.0712890625, | |
| "learning_rate": 0.00012059786187410984, | |
| "loss": 1.9446, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "eval_loss": 1.9901797771453857, | |
| "eval_runtime": 20.9042, | |
| "eval_samples_per_second": 2.392, | |
| "eval_steps_per_second": 0.335, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9159663865546217, | |
| "grad_norm": 0.07373046875, | |
| "learning_rate": 0.0001176890275122573, | |
| "loss": 1.8896, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.949579831932773, | |
| "grad_norm": 0.07373046875, | |
| "learning_rate": 0.00011476465640024814, | |
| "loss": 1.9352, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.9831932773109244, | |
| "grad_norm": 0.068359375, | |
| "learning_rate": 0.00011182731709213659, | |
| "loss": 1.8319, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.0168067226890756, | |
| "grad_norm": 0.076171875, | |
| "learning_rate": 0.00010887958953229349, | |
| "loss": 1.9105, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0504201680672267, | |
| "grad_norm": 0.0751953125, | |
| "learning_rate": 0.00010592406278937144, | |
| "loss": 2.0352, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.0840336134453783, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.00010296333278225599, | |
| "loss": 1.9011, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.068359375, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9055, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "grad_norm": 0.0654296875, | |
| "learning_rate": 9.703666721774402e-05, | |
| "loss": 1.9023, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "eval_loss": 1.9825478792190552, | |
| "eval_runtime": 20.8366, | |
| "eval_samples_per_second": 2.4, | |
| "eval_steps_per_second": 0.336, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.184873949579832, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 9.407593721062859e-05, | |
| "loss": 1.9397, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2184873949579833, | |
| "grad_norm": 0.0732421875, | |
| "learning_rate": 9.112041046770653e-05, | |
| "loss": 1.9439, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.2521008403361344, | |
| "grad_norm": 0.06982421875, | |
| "learning_rate": 8.817268290786343e-05, | |
| "loss": 1.871, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.07080078125, | |
| "learning_rate": 8.523534359975189e-05, | |
| "loss": 1.9338, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.3193277310924367, | |
| "grad_norm": 0.06884765625, | |
| "learning_rate": 8.231097248774274e-05, | |
| "loss": 1.9644, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.06591796875, | |
| "learning_rate": 7.940213812589018e-05, | |
| "loss": 1.8381, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.3865546218487395, | |
| "grad_norm": 0.07421875, | |
| "learning_rate": 7.651139542190164e-05, | |
| "loss": 1.8596, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 7.364128339309326e-05, | |
| "loss": 1.834, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "eval_loss": 1.9764043092727661, | |
| "eval_runtime": 20.8236, | |
| "eval_samples_per_second": 2.401, | |
| "eval_steps_per_second": 0.336, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.453781512605042, | |
| "grad_norm": 0.06982421875, | |
| "learning_rate": 7.079432293630244e-05, | |
| "loss": 1.8935, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.4873949579831933, | |
| "grad_norm": 0.0888671875, | |
| "learning_rate": 6.797301461371625e-05, | |
| "loss": 1.8493, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.5210084033613445, | |
| "grad_norm": 0.0712890625, | |
| "learning_rate": 6.517983645656014e-05, | |
| "loss": 1.7431, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5546218487394956, | |
| "grad_norm": 0.0751953125, | |
| "learning_rate": 6.24172417885762e-05, | |
| "loss": 1.9838, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 5.96876570712028e-05, | |
| "loss": 1.8481, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.6218487394957983, | |
| "grad_norm": 0.08056640625, | |
| "learning_rate": 5.699347977234799e-05, | |
| "loss": 1.8767, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.6554621848739495, | |
| "grad_norm": 0.07373046875, | |
| "learning_rate": 5.43370762606287e-05, | |
| "loss": 1.9405, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "grad_norm": 0.12060546875, | |
| "learning_rate": 5.172077972692553e-05, | |
| "loss": 1.9834, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "eval_loss": 1.9720275402069092, | |
| "eval_runtime": 20.9083, | |
| "eval_samples_per_second": 2.391, | |
| "eval_steps_per_second": 0.335, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.722689075630252, | |
| "grad_norm": 0.07275390625, | |
| "learning_rate": 4.914688813507797e-05, | |
| "loss": 1.9549, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.7563025210084033, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 4.661766220352097e-05, | |
| "loss": 1.9516, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.7899159663865545, | |
| "grad_norm": 0.06884765625, | |
| "learning_rate": 4.4135323419634766e-05, | |
| "loss": 1.8236, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 4.170205208855281e-05, | |
| "loss": 1.8927, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.076171875, | |
| "learning_rate": 3.931998541814069e-05, | |
| "loss": 1.8204, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.8907563025210083, | |
| "grad_norm": 0.07421875, | |
| "learning_rate": 3.69912156418289e-05, | |
| "loss": 1.9341, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.92436974789916, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 3.471778818094785e-05, | |
| "loss": 1.8912, | |
| "step": 87 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9933870538948608.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |