| { | |
| "best_metric": 2.191589117050171, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Meta-Llama-3-8B_pct_default/checkpoint-384", | |
| "epoch": 0.9887351142581268, | |
| "eval_steps": 8, | |
| "global_step": 384, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002574831026713872, | |
| "grad_norm": 7.236040115356445, | |
| "learning_rate": 3.75e-05, | |
| "loss": 2.3601, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010299324106855488, | |
| "grad_norm": 8.456689834594727, | |
| "learning_rate": 0.00015, | |
| "loss": 2.331, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020598648213710977, | |
| "grad_norm": 6.369657516479492, | |
| "learning_rate": 0.0003, | |
| "loss": 2.2535, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020598648213710977, | |
| "eval_loss": 2.273538827896118, | |
| "eval_runtime": 10.4749, | |
| "eval_samples_per_second": 23.389, | |
| "eval_steps_per_second": 2.959, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03089797232056646, | |
| "grad_norm": 11.503490447998047, | |
| "learning_rate": 0.0002999179886011389, | |
| "loss": 2.2455, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04119729642742195, | |
| "grad_norm": 7.761600494384766, | |
| "learning_rate": 0.00029967204408281613, | |
| "loss": 2.2839, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04119729642742195, | |
| "eval_loss": 2.2722208499908447, | |
| "eval_runtime": 10.4676, | |
| "eval_samples_per_second": 23.405, | |
| "eval_steps_per_second": 2.962, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.051496620534277435, | |
| "grad_norm": 8.975590705871582, | |
| "learning_rate": 0.0002992624353817517, | |
| "loss": 2.2622, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06179594464113292, | |
| "grad_norm": 12.953361511230469, | |
| "learning_rate": 0.00029868961039904624, | |
| "loss": 2.2318, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06179594464113292, | |
| "eval_loss": 2.27947735786438, | |
| "eval_runtime": 10.4821, | |
| "eval_samples_per_second": 23.373, | |
| "eval_steps_per_second": 2.957, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07209526874798841, | |
| "grad_norm": 10.436603546142578, | |
| "learning_rate": 0.00029795419551040833, | |
| "loss": 2.2613, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0823945928548439, | |
| "grad_norm": 18.31772232055664, | |
| "learning_rate": 0.0002970569948812214, | |
| "loss": 2.3022, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0823945928548439, | |
| "eval_loss": 2.272110939025879, | |
| "eval_runtime": 10.4777, | |
| "eval_samples_per_second": 23.383, | |
| "eval_steps_per_second": 2.959, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09269391696169939, | |
| "grad_norm": 15.171878814697266, | |
| "learning_rate": 0.0002959989895872009, | |
| "loss": 2.2609, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10299324106855487, | |
| "grad_norm": 10.8043794631958, | |
| "learning_rate": 0.0002947813365416023, | |
| "loss": 2.2843, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10299324106855487, | |
| "eval_loss": 2.2608468532562256, | |
| "eval_runtime": 10.4804, | |
| "eval_samples_per_second": 23.377, | |
| "eval_steps_per_second": 2.958, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11329256517541036, | |
| "grad_norm": 8.218811988830566, | |
| "learning_rate": 0.0002934053672301536, | |
| "loss": 2.2849, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12359188928226585, | |
| "grad_norm": 9.294170379638672, | |
| "learning_rate": 0.00029187258625509513, | |
| "loss": 2.2433, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12359188928226585, | |
| "eval_loss": 2.2513418197631836, | |
| "eval_runtime": 10.4749, | |
| "eval_samples_per_second": 23.389, | |
| "eval_steps_per_second": 2.959, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13389121338912133, | |
| "grad_norm": 5.276401996612549, | |
| "learning_rate": 0.0002901846696899191, | |
| "loss": 2.3066, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14419053749597682, | |
| "grad_norm": 5.453143119812012, | |
| "learning_rate": 0.0002883434632466077, | |
| "loss": 2.2617, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14419053749597682, | |
| "eval_loss": 2.2631678581237793, | |
| "eval_runtime": 10.4651, | |
| "eval_samples_per_second": 23.411, | |
| "eval_steps_per_second": 2.962, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15448986160283232, | |
| "grad_norm": 4.665058135986328, | |
| "learning_rate": 0.00028635098025737434, | |
| "loss": 2.2787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1647891857096878, | |
| "grad_norm": 5.689172267913818, | |
| "learning_rate": 0.0002842093994731145, | |
| "loss": 2.2962, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1647891857096878, | |
| "eval_loss": 2.276745080947876, | |
| "eval_runtime": 10.4548, | |
| "eval_samples_per_second": 23.434, | |
| "eval_steps_per_second": 2.965, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17508850981654328, | |
| "grad_norm": 5.17114782333374, | |
| "learning_rate": 0.00028192106268097334, | |
| "loss": 2.3209, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18538783392339878, | |
| "grad_norm": 7.912550926208496, | |
| "learning_rate": 0.0002794884721436361, | |
| "loss": 2.2573, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18538783392339878, | |
| "eval_loss": 2.2800114154815674, | |
| "eval_runtime": 10.441, | |
| "eval_samples_per_second": 23.465, | |
| "eval_steps_per_second": 2.969, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19568715803025427, | |
| "grad_norm": 7.60773229598999, | |
| "learning_rate": 0.0002769142878631403, | |
| "loss": 2.3339, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20598648213710974, | |
| "grad_norm": 4.8477959632873535, | |
| "learning_rate": 0.000274201324672203, | |
| "loss": 2.2856, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20598648213710974, | |
| "eval_loss": 2.2750115394592285, | |
| "eval_runtime": 10.4164, | |
| "eval_samples_per_second": 23.521, | |
| "eval_steps_per_second": 2.976, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21628580624396523, | |
| "grad_norm": 6.659757614135742, | |
| "learning_rate": 0.0002713525491562421, | |
| "loss": 2.2837, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22658513035082073, | |
| "grad_norm": 6.980467319488525, | |
| "learning_rate": 0.00026837107640945905, | |
| "loss": 2.3158, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22658513035082073, | |
| "eval_loss": 2.2799086570739746, | |
| "eval_runtime": 10.3932, | |
| "eval_samples_per_second": 23.573, | |
| "eval_steps_per_second": 2.983, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23688445445767622, | |
| "grad_norm": 5.191193103790283, | |
| "learning_rate": 0.00026526016662852886, | |
| "loss": 2.301, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2471837785645317, | |
| "grad_norm": 5.624630451202393, | |
| "learning_rate": 0.0002620232215476231, | |
| "loss": 2.3622, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2471837785645317, | |
| "eval_loss": 2.2860019207000732, | |
| "eval_runtime": 10.3766, | |
| "eval_samples_per_second": 23.611, | |
| "eval_steps_per_second": 2.987, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2574831026713872, | |
| "grad_norm": 5.981316089630127, | |
| "learning_rate": 0.00025866378071866334, | |
| "loss": 2.3561, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26778242677824265, | |
| "grad_norm": 6.887294292449951, | |
| "learning_rate": 0.00025518551764087326, | |
| "loss": 2.357, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.26778242677824265, | |
| "eval_loss": 2.290139675140381, | |
| "eval_runtime": 10.3578, | |
| "eval_samples_per_second": 23.654, | |
| "eval_steps_per_second": 2.993, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2780817508850982, | |
| "grad_norm": 6.06417179107666, | |
| "learning_rate": 0.00025159223574386114, | |
| "loss": 2.3055, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28838107499195365, | |
| "grad_norm": 6.5543928146362305, | |
| "learning_rate": 0.00024788786422862526, | |
| "loss": 2.3124, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28838107499195365, | |
| "eval_loss": 2.2984819412231445, | |
| "eval_runtime": 44.4261, | |
| "eval_samples_per_second": 5.515, | |
| "eval_steps_per_second": 0.698, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29868039909880917, | |
| "grad_norm": 5.269571304321289, | |
| "learning_rate": 0.00024407645377103054, | |
| "loss": 2.3163, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30897972320566464, | |
| "grad_norm": 5.49562931060791, | |
| "learning_rate": 0.00024016217209245374, | |
| "loss": 2.3646, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30897972320566464, | |
| "eval_loss": 2.2943296432495117, | |
| "eval_runtime": 44.1467, | |
| "eval_samples_per_second": 5.55, | |
| "eval_steps_per_second": 0.702, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3192790473125201, | |
| "grad_norm": 7.95536994934082, | |
| "learning_rate": 0.0002361492994024415, | |
| "loss": 2.3003, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3295783714193756, | |
| "grad_norm": 5.131826400756836, | |
| "learning_rate": 0.00023204222371836405, | |
| "loss": 2.3591, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3295783714193756, | |
| "eval_loss": 2.2891480922698975, | |
| "eval_runtime": 44.0608, | |
| "eval_samples_per_second": 5.56, | |
| "eval_steps_per_second": 0.704, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3398776955262311, | |
| "grad_norm": 6.059808731079102, | |
| "learning_rate": 0.00022784543606718227, | |
| "loss": 2.3418, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.35017701963308656, | |
| "grad_norm": 6.209086894989014, | |
| "learning_rate": 0.0002235635255745762, | |
| "loss": 2.3085, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.35017701963308656, | |
| "eval_loss": 2.2922980785369873, | |
| "eval_runtime": 44.0389, | |
| "eval_samples_per_second": 5.563, | |
| "eval_steps_per_second": 0.704, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3604763437399421, | |
| "grad_norm": 5.690346717834473, | |
| "learning_rate": 0.00021920117444680317, | |
| "loss": 2.3863, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37077566784679755, | |
| "grad_norm": 4.371445178985596, | |
| "learning_rate": 0.0002147631528507739, | |
| "loss": 2.3054, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37077566784679755, | |
| "eval_loss": 2.287841558456421, | |
| "eval_runtime": 44.2544, | |
| "eval_samples_per_second": 5.536, | |
| "eval_steps_per_second": 0.7, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.381074991953653, | |
| "grad_norm": 6.41508674621582, | |
| "learning_rate": 0.0002102543136979454, | |
| "loss": 2.2997, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39137431606050854, | |
| "grad_norm": 4.227716445922852, | |
| "learning_rate": 0.0002056795873377331, | |
| "loss": 2.3203, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39137431606050854, | |
| "eval_loss": 2.282947063446045, | |
| "eval_runtime": 44.1342, | |
| "eval_samples_per_second": 5.551, | |
| "eval_steps_per_second": 0.702, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.401673640167364, | |
| "grad_norm": 4.451204776763916, | |
| "learning_rate": 0.00020104397616624645, | |
| "loss": 2.2989, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4119729642742195, | |
| "grad_norm": 5.805531978607178, | |
| "learning_rate": 0.0001963525491562421, | |
| "loss": 2.2995, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4119729642742195, | |
| "eval_loss": 2.2782602310180664, | |
| "eval_runtime": 44.9558, | |
| "eval_samples_per_second": 5.45, | |
| "eval_steps_per_second": 0.69, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.422272288381075, | |
| "grad_norm": 4.269804000854492, | |
| "learning_rate": 0.00019161043631427666, | |
| "loss": 2.2861, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43257161248793047, | |
| "grad_norm": 4.4612135887146, | |
| "learning_rate": 0.00018682282307111987, | |
| "loss": 2.356, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43257161248793047, | |
| "eval_loss": 2.275855779647827, | |
| "eval_runtime": 10.4715, | |
| "eval_samples_per_second": 23.397, | |
| "eval_steps_per_second": 2.96, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.442870936594786, | |
| "grad_norm": 5.098966598510742, | |
| "learning_rate": 0.00018199494461156203, | |
| "loss": 2.3096, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45317026070164146, | |
| "grad_norm": 5.2250871658325195, | |
| "learning_rate": 0.00017713208014981648, | |
| "loss": 2.2942, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45317026070164146, | |
| "eval_loss": 2.2719717025756836, | |
| "eval_runtime": 10.4779, | |
| "eval_samples_per_second": 23.383, | |
| "eval_steps_per_second": 2.959, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4634695848084969, | |
| "grad_norm": 4.393130302429199, | |
| "learning_rate": 0.00017223954715677627, | |
| "loss": 2.3204, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47376890891535245, | |
| "grad_norm": 3.999812602996826, | |
| "learning_rate": 0.00016732269554543794, | |
| "loss": 2.2987, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47376890891535245, | |
| "eval_loss": 2.2650279998779297, | |
| "eval_runtime": 10.4729, | |
| "eval_samples_per_second": 23.394, | |
| "eval_steps_per_second": 2.96, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4840682330222079, | |
| "grad_norm": 4.32090425491333, | |
| "learning_rate": 0.00016238690182084986, | |
| "loss": 2.2684, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4943675571290634, | |
| "grad_norm": 4.725154876708984, | |
| "learning_rate": 0.00015743756320098332, | |
| "loss": 2.3025, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4943675571290634, | |
| "eval_loss": 2.2645103931427, | |
| "eval_runtime": 10.4773, | |
| "eval_samples_per_second": 23.384, | |
| "eval_steps_per_second": 2.959, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5046668812359189, | |
| "grad_norm": 4.0945892333984375, | |
| "learning_rate": 0.00015248009171495378, | |
| "loss": 2.2888, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5149662053427744, | |
| "grad_norm": 4.477442264556885, | |
| "learning_rate": 0.00014751990828504622, | |
| "loss": 2.294, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5149662053427744, | |
| "eval_loss": 2.2624104022979736, | |
| "eval_runtime": 10.4821, | |
| "eval_samples_per_second": 23.373, | |
| "eval_steps_per_second": 2.957, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5252655294496299, | |
| "grad_norm": 3.748276710510254, | |
| "learning_rate": 0.00014256243679901663, | |
| "loss": 2.2632, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5355648535564853, | |
| "grad_norm": 4.243557453155518, | |
| "learning_rate": 0.00013761309817915014, | |
| "loss": 2.2959, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5355648535564853, | |
| "eval_loss": 2.2678449153900146, | |
| "eval_runtime": 10.4697, | |
| "eval_samples_per_second": 23.401, | |
| "eval_steps_per_second": 2.961, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5458641776633408, | |
| "grad_norm": 3.7038612365722656, | |
| "learning_rate": 0.00013267730445456208, | |
| "loss": 2.2426, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5561635017701964, | |
| "grad_norm": 5.393055438995361, | |
| "learning_rate": 0.00012776045284322368, | |
| "loss": 2.3074, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5561635017701964, | |
| "eval_loss": 2.2524540424346924, | |
| "eval_runtime": 10.4594, | |
| "eval_samples_per_second": 23.424, | |
| "eval_steps_per_second": 2.964, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5664628258770518, | |
| "grad_norm": 5.647940635681152, | |
| "learning_rate": 0.00012286791985018355, | |
| "loss": 2.2712, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5767621499839073, | |
| "grad_norm": 3.919537305831909, | |
| "learning_rate": 0.00011800505538843798, | |
| "loss": 2.2862, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5767621499839073, | |
| "eval_loss": 2.2530322074890137, | |
| "eval_runtime": 10.464, | |
| "eval_samples_per_second": 23.414, | |
| "eval_steps_per_second": 2.963, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5870614740907628, | |
| "grad_norm": 4.026923656463623, | |
| "learning_rate": 0.00011317717692888012, | |
| "loss": 2.2933, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5973607981976183, | |
| "grad_norm": 4.218359470367432, | |
| "learning_rate": 0.00010838956368572334, | |
| "loss": 2.2745, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5973607981976183, | |
| "eval_loss": 2.2494354248046875, | |
| "eval_runtime": 10.4537, | |
| "eval_samples_per_second": 23.437, | |
| "eval_steps_per_second": 2.965, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6076601223044737, | |
| "grad_norm": 3.6584322452545166, | |
| "learning_rate": 0.0001036474508437579, | |
| "loss": 2.3179, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6179594464113293, | |
| "grad_norm": 3.86560320854187, | |
| "learning_rate": 9.895602383375353e-05, | |
| "loss": 2.2422, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6179594464113293, | |
| "eval_loss": 2.239821434020996, | |
| "eval_runtime": 10.4187, | |
| "eval_samples_per_second": 23.515, | |
| "eval_steps_per_second": 2.975, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6282587705181848, | |
| "grad_norm": 6.043626308441162, | |
| "learning_rate": 9.432041266226686e-05, | |
| "loss": 2.2811, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6385580946250402, | |
| "grad_norm": 4.4510393142700195, | |
| "learning_rate": 8.97456863020546e-05, | |
| "loss": 2.275, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6385580946250402, | |
| "eval_loss": 2.2399165630340576, | |
| "eval_runtime": 10.4081, | |
| "eval_samples_per_second": 23.539, | |
| "eval_steps_per_second": 2.978, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6488574187318957, | |
| "grad_norm": 3.6573245525360107, | |
| "learning_rate": 8.523684714922608e-05, | |
| "loss": 2.2348, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6591567428387513, | |
| "grad_norm": 4.474267482757568, | |
| "learning_rate": 8.079882555319684e-05, | |
| "loss": 2.2632, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6591567428387513, | |
| "eval_loss": 2.2398147583007812, | |
| "eval_runtime": 10.3617, | |
| "eval_samples_per_second": 23.645, | |
| "eval_steps_per_second": 2.992, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6694560669456067, | |
| "grad_norm": 4.0849995613098145, | |
| "learning_rate": 7.643647442542382e-05, | |
| "loss": 2.2624, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6797553910524622, | |
| "grad_norm": 3.5736443996429443, | |
| "learning_rate": 7.215456393281776e-05, | |
| "loss": 2.2198, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6797553910524622, | |
| "eval_loss": 2.2288267612457275, | |
| "eval_runtime": 10.3503, | |
| "eval_samples_per_second": 23.671, | |
| "eval_steps_per_second": 2.995, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6900547151593177, | |
| "grad_norm": 3.700812816619873, | |
| "learning_rate": 6.795777628163599e-05, | |
| "loss": 2.2498, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7003540392661731, | |
| "grad_norm": 3.252271890640259, | |
| "learning_rate": 6.385070059755846e-05, | |
| "loss": 2.2732, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7003540392661731, | |
| "eval_loss": 2.2233266830444336, | |
| "eval_runtime": 44.1123, | |
| "eval_samples_per_second": 5.554, | |
| "eval_steps_per_second": 0.703, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7106533633730286, | |
| "grad_norm": 3.207204818725586, | |
| "learning_rate": 5.983782790754623e-05, | |
| "loss": 2.2661, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7209526874798842, | |
| "grad_norm": 3.933907985687256, | |
| "learning_rate": 5.592354622896944e-05, | |
| "loss": 2.2576, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7209526874798842, | |
| "eval_loss": 2.2177813053131104, | |
| "eval_runtime": 45.0771, | |
| "eval_samples_per_second": 5.435, | |
| "eval_steps_per_second": 0.688, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7312520115867396, | |
| "grad_norm": 3.623133659362793, | |
| "learning_rate": 5.211213577137469e-05, | |
| "loss": 2.253, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7415513356935951, | |
| "grad_norm": 3.6505510807037354, | |
| "learning_rate": 4.840776425613886e-05, | |
| "loss": 2.2606, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7415513356935951, | |
| "eval_loss": 2.2098140716552734, | |
| "eval_runtime": 44.4469, | |
| "eval_samples_per_second": 5.512, | |
| "eval_steps_per_second": 0.697, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7518506598004506, | |
| "grad_norm": 3.72786021232605, | |
| "learning_rate": 4.481448235912671e-05, | |
| "loss": 2.2492, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.762149983907306, | |
| "grad_norm": 3.0990045070648193, | |
| "learning_rate": 4.133621928133665e-05, | |
| "loss": 2.2559, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.762149983907306, | |
| "eval_loss": 2.2151412963867188, | |
| "eval_runtime": 44.8388, | |
| "eval_samples_per_second": 5.464, | |
| "eval_steps_per_second": 0.691, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7724493080141616, | |
| "grad_norm": 3.5728273391723633, | |
| "learning_rate": 3.797677845237696e-05, | |
| "loss": 2.2018, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7827486321210171, | |
| "grad_norm": 3.1590304374694824, | |
| "learning_rate": 3.473983337147118e-05, | |
| "loss": 2.2852, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7827486321210171, | |
| "eval_loss": 2.2048428058624268, | |
| "eval_runtime": 45.1815, | |
| "eval_samples_per_second": 5.423, | |
| "eval_steps_per_second": 0.686, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7930479562278725, | |
| "grad_norm": 3.989384412765503, | |
| "learning_rate": 3.162892359054098e-05, | |
| "loss": 2.2077, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.803347280334728, | |
| "grad_norm": 4.393310070037842, | |
| "learning_rate": 2.8647450843757897e-05, | |
| "loss": 2.2252, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.803347280334728, | |
| "eval_loss": 2.2025868892669678, | |
| "eval_runtime": 45.7427, | |
| "eval_samples_per_second": 5.356, | |
| "eval_steps_per_second": 0.678, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8136466044415835, | |
| "grad_norm": 3.008932590484619, | |
| "learning_rate": 2.5798675327796993e-05, | |
| "loss": 2.1846, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.823945928548439, | |
| "grad_norm": 3.083261251449585, | |
| "learning_rate": 2.3085712136859668e-05, | |
| "loss": 2.2024, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.823945928548439, | |
| "eval_loss": 2.2029316425323486, | |
| "eval_runtime": 44.6941, | |
| "eval_samples_per_second": 5.482, | |
| "eval_steps_per_second": 0.694, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8342452526552945, | |
| "grad_norm": 3.446568250656128, | |
| "learning_rate": 2.0511527856363912e-05, | |
| "loss": 2.2635, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.84454457676215, | |
| "grad_norm": 3.727339029312134, | |
| "learning_rate": 1.8078937319026654e-05, | |
| "loss": 2.2339, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.84454457676215, | |
| "eval_loss": 2.196918249130249, | |
| "eval_runtime": 10.48, | |
| "eval_samples_per_second": 23.378, | |
| "eval_steps_per_second": 2.958, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8548439008690055, | |
| "grad_norm": 3.6961987018585205, | |
| "learning_rate": 1.579060052688548e-05, | |
| "loss": 2.21, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8651432249758609, | |
| "grad_norm": 2.9594602584838867, | |
| "learning_rate": 1.3649019742625623e-05, | |
| "loss": 2.2468, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8651432249758609, | |
| "eval_loss": 2.197859048843384, | |
| "eval_runtime": 10.4828, | |
| "eval_samples_per_second": 23.372, | |
| "eval_steps_per_second": 2.957, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8754425490827165, | |
| "grad_norm": 3.7084383964538574, | |
| "learning_rate": 1.1656536753392287e-05, | |
| "loss": 2.2378, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.885741873189572, | |
| "grad_norm": 4.392552852630615, | |
| "learning_rate": 9.815330310080887e-06, | |
| "loss": 2.2582, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.885741873189572, | |
| "eval_loss": 2.1931886672973633, | |
| "eval_runtime": 10.4854, | |
| "eval_samples_per_second": 23.366, | |
| "eval_steps_per_second": 2.956, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8960411972964274, | |
| "grad_norm": 3.1651930809020996, | |
| "learning_rate": 8.127413744904804e-06, | |
| "loss": 2.2329, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9063405214032829, | |
| "grad_norm": 3.342674493789673, | |
| "learning_rate": 6.594632769846353e-06, | |
| "loss": 2.223, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9063405214032829, | |
| "eval_loss": 2.1925430297851562, | |
| "eval_runtime": 10.4791, | |
| "eval_samples_per_second": 23.38, | |
| "eval_steps_per_second": 2.958, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9166398455101384, | |
| "grad_norm": 3.3589231967926025, | |
| "learning_rate": 5.218663458397715e-06, | |
| "loss": 2.2147, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9269391696169939, | |
| "grad_norm": 2.758237361907959, | |
| "learning_rate": 4.001010412799138e-06, | |
| "loss": 2.1887, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9269391696169939, | |
| "eval_loss": 2.1936519145965576, | |
| "eval_runtime": 10.4939, | |
| "eval_samples_per_second": 23.347, | |
| "eval_steps_per_second": 2.954, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9372384937238494, | |
| "grad_norm": 3.3825769424438477, | |
| "learning_rate": 2.9430051187785962e-06, | |
| "loss": 2.2439, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9475378178307049, | |
| "grad_norm": 3.4175548553466797, | |
| "learning_rate": 2.0458044895916513e-06, | |
| "loss": 2.218, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9475378178307049, | |
| "eval_loss": 2.1924386024475098, | |
| "eval_runtime": 10.4759, | |
| "eval_samples_per_second": 23.387, | |
| "eval_steps_per_second": 2.959, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9578371419375603, | |
| "grad_norm": 3.223475217819214, | |
| "learning_rate": 1.3103896009537207e-06, | |
| "loss": 2.195, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9681364660444158, | |
| "grad_norm": 2.9230334758758545, | |
| "learning_rate": 7.375646182482875e-07, | |
| "loss": 2.258, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9681364660444158, | |
| "eval_loss": 2.1916966438293457, | |
| "eval_runtime": 10.4626, | |
| "eval_samples_per_second": 23.417, | |
| "eval_steps_per_second": 2.963, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9784357901512714, | |
| "grad_norm": 4.080029487609863, | |
| "learning_rate": 3.2795591718381975e-07, | |
| "loss": 2.1991, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9887351142581268, | |
| "grad_norm": 2.951921224594116, | |
| "learning_rate": 8.201139886109264e-08, | |
| "loss": 2.2479, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9887351142581268, | |
| "eval_loss": 2.191589117050171, | |
| "eval_runtime": 10.4621, | |
| "eval_samples_per_second": 23.418, | |
| "eval_steps_per_second": 2.963, | |
| "step": 384 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 388, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.99888145270571e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |