| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9970326409495549, | |
| "eval_steps": 42, | |
| "global_step": 168, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.0981353223323822, | |
| "learning_rate": 2e-05, | |
| "loss": 0.641, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.6416735053062439, | |
| "eval_runtime": 21.4326, | |
| "eval_samples_per_second": 46.331, | |
| "eval_steps_per_second": 11.618, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.09748291969299316, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6396, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.09947647899389267, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6397, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.09976381808519363, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6371, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.10493721067905426, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6491, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.1144007071852684, | |
| "learning_rate": 0.00012, | |
| "loss": 0.6218, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.08536222577095032, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6177, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.11926598846912384, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5861, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.15648387372493744, | |
| "learning_rate": 0.00018, | |
| "loss": 0.6006, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.12172720581293106, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5845, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.09348208457231522, | |
| "learning_rate": 0.0001999988739622358, | |
| "loss": 0.5471, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.07471276819705963, | |
| "learning_rate": 0.00019999549587430254, | |
| "loss": 0.578, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.07200929522514343, | |
| "learning_rate": 0.00019998986581227718, | |
| "loss": 0.5328, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.07460763305425644, | |
| "learning_rate": 0.000199981983902953, | |
| "loss": 0.5651, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.07441641390323639, | |
| "learning_rate": 0.00019997185032383664, | |
| "loss": 0.5589, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.07513019442558289, | |
| "learning_rate": 0.00019995946530314385, | |
| "loss": 0.5736, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.06902395933866501, | |
| "learning_rate": 0.00019994482911979468, | |
| "loss": 0.556, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.07314619421958923, | |
| "learning_rate": 0.00019992794210340706, | |
| "loss": 0.5469, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.06833848357200623, | |
| "learning_rate": 0.00019990880463428937, | |
| "loss": 0.5448, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.07301248610019684, | |
| "learning_rate": 0.00019988741714343177, | |
| "loss": 0.5612, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.07063400000333786, | |
| "learning_rate": 0.0001998637801124968, | |
| "loss": 0.5446, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.06935883313417435, | |
| "learning_rate": 0.00019983789407380828, | |
| "loss": 0.5223, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.06576420366764069, | |
| "learning_rate": 0.00019980975961033924, | |
| "loss": 0.5351, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.07276671379804611, | |
| "learning_rate": 0.00019977937735569915, | |
| "loss": 0.5423, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.0756976306438446, | |
| "learning_rate": 0.00019974674799411925, | |
| "loss": 0.5344, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.06945928931236267, | |
| "learning_rate": 0.00019971187226043745, | |
| "loss": 0.5198, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.06587155908346176, | |
| "learning_rate": 0.0001996747509400816, | |
| "loss": 0.5175, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.0752682164311409, | |
| "learning_rate": 0.0001996353848690519, | |
| "loss": 0.5068, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.0740601122379303, | |
| "learning_rate": 0.00019959377493390196, | |
| "loss": 0.535, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.07076304405927658, | |
| "learning_rate": 0.00019954992207171898, | |
| "loss": 0.5079, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.0776033028960228, | |
| "learning_rate": 0.00019950382727010254, | |
| "loss": 0.5124, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.0779872015118599, | |
| "learning_rate": 0.00019945549156714234, | |
| "loss": 0.5146, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.08037945628166199, | |
| "learning_rate": 0.00019940491605139498, | |
| "loss": 0.5189, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.06880298256874084, | |
| "learning_rate": 0.0001993521018618592, | |
| "loss": 0.506, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.0755767747759819, | |
| "learning_rate": 0.00019929705018795053, | |
| "loss": 0.4997, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.07505559921264648, | |
| "learning_rate": 0.00019923976226947417, | |
| "loss": 0.502, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.07533205300569534, | |
| "learning_rate": 0.00019918023939659733, | |
| "loss": 0.5093, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.0748637244105339, | |
| "learning_rate": 0.0001991184829098201, | |
| "loss": 0.4976, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.076931431889534, | |
| "learning_rate": 0.00019905449419994518, | |
| "loss": 0.4992, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.07511387020349503, | |
| "learning_rate": 0.0001989882747080466, | |
| "loss": 0.5069, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.0723625123500824, | |
| "learning_rate": 0.00019891982592543746, | |
| "loss": 0.4952, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.07320375740528107, | |
| "learning_rate": 0.00019884914939363588, | |
| "loss": 0.5093, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.5259941220283508, | |
| "eval_runtime": 21.4684, | |
| "eval_samples_per_second": 46.254, | |
| "eval_steps_per_second": 11.598, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.07251272350549698, | |
| "learning_rate": 0.00019877624670433086, | |
| "loss": 0.4931, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.07731667906045914, | |
| "learning_rate": 0.00019870111949934599, | |
| "loss": 0.4879, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.074358269572258, | |
| "learning_rate": 0.00019862376947060264, | |
| "loss": 0.5049, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.0808371901512146, | |
| "learning_rate": 0.0001985441983600819, | |
| "loss": 0.517, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.07559769600629807, | |
| "learning_rate": 0.00019846240795978528, | |
| "loss": 0.4834, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.07425505667924881, | |
| "learning_rate": 0.00019837840011169438, | |
| "loss": 0.5138, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.07782939821481705, | |
| "learning_rate": 0.00019829217670772935, | |
| "loss": 0.4858, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.0754002034664154, | |
| "learning_rate": 0.00019820373968970642, | |
| "loss": 0.4941, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.07364428788423538, | |
| "learning_rate": 0.000198113091049294, | |
| "loss": 0.4835, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.08309967815876007, | |
| "learning_rate": 0.00019802023282796796, | |
| "loss": 0.5237, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.07548778504133224, | |
| "learning_rate": 0.00019792516711696556, | |
| "loss": 0.4923, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.07607278972864151, | |
| "learning_rate": 0.0001978278960572384, | |
| "loss": 0.4971, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.07432844489812851, | |
| "learning_rate": 0.00019772842183940422, | |
| "loss": 0.4874, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.077260322868824, | |
| "learning_rate": 0.00019762674670369755, | |
| "loss": 0.5067, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.08594146370887756, | |
| "learning_rate": 0.00019752287293991927, | |
| "loss": 0.4804, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.075816310942173, | |
| "learning_rate": 0.00019741680288738492, | |
| "loss": 0.4738, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.07784326374530792, | |
| "learning_rate": 0.00019730853893487228, | |
| "loss": 0.4768, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.08903329074382782, | |
| "learning_rate": 0.00019719808352056724, | |
| "loss": 0.4773, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.07911587506532669, | |
| "learning_rate": 0.00019708543913200924, | |
| "loss": 0.4672, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.07881385087966919, | |
| "learning_rate": 0.00019697060830603494, | |
| "loss": 0.4824, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.08292945474386215, | |
| "learning_rate": 0.00019685359362872125, | |
| "loss": 0.4814, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.08237861096858978, | |
| "learning_rate": 0.00019673439773532713, | |
| "loss": 0.486, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.07958442717790604, | |
| "learning_rate": 0.0001966130233102341, | |
| "loss": 0.4913, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.07969169318675995, | |
| "learning_rate": 0.00019648947308688593, | |
| "loss": 0.4781, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.08310563862323761, | |
| "learning_rate": 0.00019636374984772692, | |
| "loss": 0.4811, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.07763976603746414, | |
| "learning_rate": 0.00019623585642413938, | |
| "loss": 0.4809, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.0927213802933693, | |
| "learning_rate": 0.00019610579569637982, | |
| "loss": 0.5019, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.08405344933271408, | |
| "learning_rate": 0.000195973570593514, | |
| "loss": 0.5001, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.07862479984760284, | |
| "learning_rate": 0.0001958391840933512, | |
| "loss": 0.4894, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.07815296947956085, | |
| "learning_rate": 0.00019570263922237687, | |
| "loss": 0.4676, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.07999672740697861, | |
| "learning_rate": 0.00019556393905568458, | |
| "loss": 0.4857, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.08266247063875198, | |
| "learning_rate": 0.0001954230867169069, | |
| "loss": 0.4842, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.08117777854204178, | |
| "learning_rate": 0.00019528008537814486, | |
| "loss": 0.4602, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.08203484117984772, | |
| "learning_rate": 0.00019513493825989664, | |
| "loss": 0.4761, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.07647153735160828, | |
| "learning_rate": 0.00019498764863098495, | |
| "loss": 0.4839, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.0811714455485344, | |
| "learning_rate": 0.00019483821980848347, | |
| "loss": 0.4803, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.08266978710889816, | |
| "learning_rate": 0.00019468665515764215, | |
| "loss": 0.4665, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.07869689911603928, | |
| "learning_rate": 0.00019453295809181143, | |
| "loss": 0.4857, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.08934654295444489, | |
| "learning_rate": 0.00019437713207236525, | |
| "loss": 0.4825, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.07842836529016495, | |
| "learning_rate": 0.00019421918060862333, | |
| "loss": 0.4609, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.08244986832141876, | |
| "learning_rate": 0.0001940591072577719, | |
| "loss": 0.4688, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.07819854468107224, | |
| "learning_rate": 0.00019389691562478374, | |
| "loss": 0.4665, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.5117939114570618, | |
| "eval_runtime": 21.4742, | |
| "eval_samples_per_second": 46.242, | |
| "eval_steps_per_second": 11.595, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.0837428942322731, | |
| "learning_rate": 0.0001937326093623369, | |
| "loss": 0.4952, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.07781701534986496, | |
| "learning_rate": 0.00019356619217073253, | |
| "loss": 0.467, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.08447270840406418, | |
| "learning_rate": 0.00019339766779781145, | |
| "loss": 0.4838, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.08231997489929199, | |
| "learning_rate": 0.00019322704003886987, | |
| "loss": 0.4611, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.08507382869720459, | |
| "learning_rate": 0.00019305431273657374, | |
| "loss": 0.4757, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.08521989732980728, | |
| "learning_rate": 0.0001928794897808724, | |
| "loss": 0.4854, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.0963786169886589, | |
| "learning_rate": 0.00019270257510891082, | |
| "loss": 0.4505, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.08671442419290543, | |
| "learning_rate": 0.0001925235727049411, | |
| "loss": 0.4766, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.09087081998586655, | |
| "learning_rate": 0.0001923424866002325, | |
| "loss": 0.4966, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.07899381965398788, | |
| "learning_rate": 0.00019215932087298092, | |
| "loss": 0.4638, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.09070860594511032, | |
| "learning_rate": 0.00019197407964821684, | |
| "loss": 0.4847, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.0885949656367302, | |
| "learning_rate": 0.00019178676709771258, | |
| "loss": 0.4648, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.09253839403390884, | |
| "learning_rate": 0.00019159738743988825, | |
| "loss": 0.459, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.08571318536996841, | |
| "learning_rate": 0.00019140594493971674, | |
| "loss": 0.4797, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.07787954807281494, | |
| "learning_rate": 0.0001912124439086278, | |
| "loss": 0.4547, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.08822935819625854, | |
| "learning_rate": 0.00019101688870441078, | |
| "loss": 0.4511, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.08409956842660904, | |
| "learning_rate": 0.0001908192837311166, | |
| "loss": 0.4631, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.08279416710138321, | |
| "learning_rate": 0.00019061963343895846, | |
| "loss": 0.4696, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.09696204960346222, | |
| "learning_rate": 0.00019041794232421176, | |
| "loss": 0.4862, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.08494329452514648, | |
| "learning_rate": 0.00019021421492911272, | |
| "loss": 0.4557, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.08702557533979416, | |
| "learning_rate": 0.00019000845584175616, | |
| "loss": 0.4693, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.09048158675432205, | |
| "learning_rate": 0.00018980066969599216, | |
| "loss": 0.4714, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.08462114632129669, | |
| "learning_rate": 0.0001895908611713216, | |
| "loss": 0.4632, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.09956546127796173, | |
| "learning_rate": 0.00018937903499279102, | |
| "loss": 0.4638, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.08630617707967758, | |
| "learning_rate": 0.00018916519593088584, | |
| "loss": 0.4499, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.08207620680332184, | |
| "learning_rate": 0.0001889493488014233, | |
| "loss": 0.4603, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.08473565429449081, | |
| "learning_rate": 0.00018873149846544376, | |
| "loss": 0.4571, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.08818928152322769, | |
| "learning_rate": 0.00018851164982910135, | |
| "loss": 0.4489, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.08116699010133743, | |
| "learning_rate": 0.00018828980784355338, | |
| "loss": 0.4578, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.08832226693630219, | |
| "learning_rate": 0.00018806597750484897, | |
| "loss": 0.4719, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.08624406903982162, | |
| "learning_rate": 0.0001878401638538163, | |
| "loss": 0.4628, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.08936543017625809, | |
| "learning_rate": 0.00018761237197594945, | |
| "loss": 0.4533, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.08579661697149277, | |
| "learning_rate": 0.00018738260700129354, | |
| "loss": 0.4772, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.08271288126707077, | |
| "learning_rate": 0.0001871508741043293, | |
| "loss": 0.4773, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.08224964886903763, | |
| "learning_rate": 0.0001869171785038566, | |
| "loss": 0.4635, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.087012380361557, | |
| "learning_rate": 0.00018668152546287686, | |
| "loss": 0.4559, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.08352699875831604, | |
| "learning_rate": 0.00018644392028847458, | |
| "loss": 0.4485, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.08281444013118744, | |
| "learning_rate": 0.00018620436833169772, | |
| "loss": 0.4393, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.08376545459032059, | |
| "learning_rate": 0.00018596287498743732, | |
| "loss": 0.4525, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.08526434749364853, | |
| "learning_rate": 0.0001857194456943058, | |
| "loss": 0.4456, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.08151934295892715, | |
| "learning_rate": 0.0001854740859345148, | |
| "loss": 0.4576, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.08793777972459793, | |
| "learning_rate": 0.0001852268012337514, | |
| "loss": 0.4431, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.5042669773101807, | |
| "eval_runtime": 21.4592, | |
| "eval_samples_per_second": 46.274, | |
| "eval_steps_per_second": 11.603, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.08135095983743668, | |
| "learning_rate": 0.00018497759716105377, | |
| "loss": 0.4384, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.0917576476931572, | |
| "learning_rate": 0.0001847264793286859, | |
| "loss": 0.4687, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.08832691609859467, | |
| "learning_rate": 0.00018447345339201102, | |
| "loss": 0.4386, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.08340886980295181, | |
| "learning_rate": 0.00018421852504936438, | |
| "loss": 0.4512, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.08589499443769455, | |
| "learning_rate": 0.00018396170004192475, | |
| "loss": 0.4387, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.08753557503223419, | |
| "learning_rate": 0.00018370298415358526, | |
| "loss": 0.4615, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.08406232297420502, | |
| "learning_rate": 0.00018344238321082315, | |
| "loss": 0.4465, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.08514856547117233, | |
| "learning_rate": 0.0001831799030825685, | |
| "loss": 0.4516, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.09259331226348877, | |
| "learning_rate": 0.000182915549680072, | |
| "loss": 0.4387, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.08862275630235672, | |
| "learning_rate": 0.00018264932895677193, | |
| "loss": 0.4434, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.08515379577875137, | |
| "learning_rate": 0.0001823812469081601, | |
| "loss": 0.4425, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.09041007608175278, | |
| "learning_rate": 0.00018211130957164668, | |
| "loss": 0.4607, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.08312032371759415, | |
| "learning_rate": 0.0001818395230264244, | |
| "loss": 0.442, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.08981412649154663, | |
| "learning_rate": 0.00018156589339333152, | |
| "loss": 0.4608, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.08991118520498276, | |
| "learning_rate": 0.00018129042683471402, | |
| "loss": 0.451, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.08628728240728378, | |
| "learning_rate": 0.00018101312955428692, | |
| "loss": 0.4453, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.08730859309434891, | |
| "learning_rate": 0.00018073400779699435, | |
| "loss": 0.4485, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.08489865809679031, | |
| "learning_rate": 0.0001804530678488691, | |
| "loss": 0.4592, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.08439410477876663, | |
| "learning_rate": 0.00018017031603689102, | |
| "loss": 0.4326, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.09346488118171692, | |
| "learning_rate": 0.0001798857587288445, | |
| "loss": 0.4484, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.09130821377038956, | |
| "learning_rate": 0.00017959940233317498, | |
| "loss": 0.4502, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.08846256881952286, | |
| "learning_rate": 0.0001793112532988448, | |
| "loss": 0.4322, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.09061886370182037, | |
| "learning_rate": 0.00017902131811518786, | |
| "loss": 0.4437, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.09259927272796631, | |
| "learning_rate": 0.00017872960331176345, | |
| "loss": 0.4545, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.09632189571857452, | |
| "learning_rate": 0.00017843611545820926, | |
| "loss": 0.4515, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.08714065700769424, | |
| "learning_rate": 0.00017814086116409348, | |
| "loss": 0.4602, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.09537078440189362, | |
| "learning_rate": 0.00017784384707876576, | |
| "loss": 0.4482, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.09175322949886322, | |
| "learning_rate": 0.00017754507989120764, | |
| "loss": 0.4681, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.08962789177894592, | |
| "learning_rate": 0.00017724456632988187, | |
| "loss": 0.4304, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.09643880277872086, | |
| "learning_rate": 0.00017694231316258077, | |
| "loss": 0.4532, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.08335065096616745, | |
| "learning_rate": 0.00017663832719627402, | |
| "loss": 0.4504, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.087184838950634, | |
| "learning_rate": 0.0001763326152769551, | |
| "loss": 0.4752, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.08858635276556015, | |
| "learning_rate": 0.0001760251842894874, | |
| "loss": 0.4413, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.08101391792297363, | |
| "learning_rate": 0.00017571604115744892, | |
| "loss": 0.4465, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.08623132854700089, | |
| "learning_rate": 0.0001754051928429765, | |
| "loss": 0.4673, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.0922100692987442, | |
| "learning_rate": 0.00017509264634660895, | |
| "loss": 0.4587, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.08243449032306671, | |
| "learning_rate": 0.00017477840870712945, | |
| "loss": 0.4368, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.0845554992556572, | |
| "learning_rate": 0.00017446248700140693, | |
| "loss": 0.4209, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.08277452737092972, | |
| "learning_rate": 0.00017414488834423687, | |
| "loss": 0.4397, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.0826331302523613, | |
| "learning_rate": 0.00017382561988818086, | |
| "loss": 0.4333, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.08441821485757828, | |
| "learning_rate": 0.0001735046888234057, | |
| "loss": 0.4496, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.08665426075458527, | |
| "learning_rate": 0.00017318210237752136, | |
| "loss": 0.4523, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.4984985589981079, | |
| "eval_runtime": 21.4662, | |
| "eval_samples_per_second": 46.259, | |
| "eval_steps_per_second": 11.6, | |
| "step": 168 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 672, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 168, | |
| "total_flos": 5.013190238546166e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |