| { | |
| "best_metric": 2.2854506969451904, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_ortho/checkpoint-8", | |
| "epoch": 0.9980657640232108, | |
| "eval_steps": 8, | |
| "global_step": 387, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025789813023855577, | |
| "grad_norm": 10.031487464904785, | |
| "learning_rate": 3.75e-05, | |
| "loss": 2.128, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010315925209542231, | |
| "grad_norm": 7.364378929138184, | |
| "learning_rate": 0.00015, | |
| "loss": 2.0612, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020631850419084462, | |
| "grad_norm": 23.759828567504883, | |
| "learning_rate": 0.0003, | |
| "loss": 2.1243, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020631850419084462, | |
| "eval_loss": 2.2854506969451904, | |
| "eval_runtime": 11.1093, | |
| "eval_samples_per_second": 22.054, | |
| "eval_steps_per_second": 2.79, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030947775628626693, | |
| "grad_norm": 133.17942810058594, | |
| "learning_rate": 0.00029991755529206284, | |
| "loss": 2.465, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.041263700838168924, | |
| "grad_norm": 254.61338806152344, | |
| "learning_rate": 0.0002996703117966496, | |
| "loss": 9.1803, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.041263700838168924, | |
| "eval_loss": 12.813275337219238, | |
| "eval_runtime": 10.7499, | |
| "eval_samples_per_second": 22.791, | |
| "eval_steps_per_second": 2.884, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05157962604771115, | |
| "grad_norm": 87.40137481689453, | |
| "learning_rate": 0.00029925854129933066, | |
| "loss": 12.8875, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.061895551257253385, | |
| "grad_norm": 22.197757720947266, | |
| "learning_rate": 0.0002986826964440844, | |
| "loss": 9.137, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.061895551257253385, | |
| "eval_loss": 8.456737518310547, | |
| "eval_runtime": 10.6838, | |
| "eval_samples_per_second": 22.932, | |
| "eval_steps_per_second": 2.902, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07221147646679561, | |
| "grad_norm": 11.230401992797852, | |
| "learning_rate": 0.00029794341023572295, | |
| "loss": 7.9575, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08252740167633785, | |
| "grad_norm": 19.19782066345215, | |
| "learning_rate": 0.0002970414953440533, | |
| "loss": 8.3645, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08252740167633785, | |
| "eval_loss": 8.26063346862793, | |
| "eval_runtime": 10.699, | |
| "eval_samples_per_second": 22.899, | |
| "eval_steps_per_second": 2.897, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09284332688588008, | |
| "grad_norm": 13.250898361206055, | |
| "learning_rate": 0.00029597794321054006, | |
| "loss": 8.0337, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1031592520954223, | |
| "grad_norm": 15.708863258361816, | |
| "learning_rate": 0.00029475392295845, | |
| "loss": 8.5251, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1031592520954223, | |
| "eval_loss": 7.756398677825928, | |
| "eval_runtime": 10.6841, | |
| "eval_samples_per_second": 22.931, | |
| "eval_steps_per_second": 2.901, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11347517730496454, | |
| "grad_norm": 8.025888442993164, | |
| "learning_rate": 0.0002933707801076791, | |
| "loss": 7.708, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12379110251450677, | |
| "grad_norm": 72.3178482055664, | |
| "learning_rate": 0.00029183003509567217, | |
| "loss": 9.4881, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12379110251450677, | |
| "eval_loss": 9.306961059570312, | |
| "eval_runtime": 10.6939, | |
| "eval_samples_per_second": 22.91, | |
| "eval_steps_per_second": 2.899, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.134107027724049, | |
| "grad_norm": 13.741963386535645, | |
| "learning_rate": 0.000290133381606063, | |
| "loss": 8.36, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14442295293359123, | |
| "grad_norm": 18.59068489074707, | |
| "learning_rate": 0.0002882826847068703, | |
| "loss": 7.7111, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14442295293359123, | |
| "eval_loss": 7.73629903793335, | |
| "eval_runtime": 10.7184, | |
| "eval_samples_per_second": 22.858, | |
| "eval_steps_per_second": 2.892, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15473887814313347, | |
| "grad_norm": 20.837697982788086, | |
| "learning_rate": 0.00028627997880029875, | |
| "loss": 7.6612, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1650548033526757, | |
| "grad_norm": 8.735236167907715, | |
| "learning_rate": 0.0002841274653863955, | |
| "loss": 7.6126, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1650548033526757, | |
| "eval_loss": 7.594820022583008, | |
| "eval_runtime": 10.664, | |
| "eval_samples_per_second": 22.974, | |
| "eval_steps_per_second": 2.907, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17537072856221791, | |
| "grad_norm": 12.856707572937012, | |
| "learning_rate": 0.00028182751064302397, | |
| "loss": 7.5332, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18568665377176016, | |
| "grad_norm": 13.578782081604004, | |
| "learning_rate": 0.0002793826428248118, | |
| "loss": 7.6789, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18568665377176016, | |
| "eval_loss": 7.6147379875183105, | |
| "eval_runtime": 10.6601, | |
| "eval_samples_per_second": 22.983, | |
| "eval_steps_per_second": 2.908, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19600257898130238, | |
| "grad_norm": 7.203093528747559, | |
| "learning_rate": 0.0002767955494839353, | |
| "loss": 7.551, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063185041908446, | |
| "grad_norm": 5.764775276184082, | |
| "learning_rate": 0.00027406907451579294, | |
| "loss": 7.7404, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063185041908446, | |
| "eval_loss": 7.632232666015625, | |
| "eval_runtime": 10.6044, | |
| "eval_samples_per_second": 23.104, | |
| "eval_steps_per_second": 2.923, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21663442940038685, | |
| "grad_norm": 15.057692527770996, | |
| "learning_rate": 0.0002712062150328175, | |
| "loss": 7.7502, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22695035460992907, | |
| "grad_norm": 7.761707305908203, | |
| "learning_rate": 0.0002682101180698615, | |
| "loss": 7.7173, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22695035460992907, | |
| "eval_loss": 7.6740193367004395, | |
| "eval_runtime": 10.575, | |
| "eval_samples_per_second": 23.168, | |
| "eval_steps_per_second": 2.931, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23726627981947132, | |
| "grad_norm": 10.52561092376709, | |
| "learning_rate": 0.000265084077124779, | |
| "loss": 7.593, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24758220502901354, | |
| "grad_norm": 13.96662425994873, | |
| "learning_rate": 0.0002618315285380063, | |
| "loss": 7.7113, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24758220502901354, | |
| "eval_loss": 7.674198627471924, | |
| "eval_runtime": 10.533, | |
| "eval_samples_per_second": 23.26, | |
| "eval_steps_per_second": 2.943, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2578981302385558, | |
| "grad_norm": 5.200709819793701, | |
| "learning_rate": 0.00025845604771512044, | |
| "loss": 7.698, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.268214055448098, | |
| "grad_norm": 5.996535301208496, | |
| "learning_rate": 0.00025496134519652946, | |
| "loss": 7.6961, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.268214055448098, | |
| "eval_loss": 7.642160415649414, | |
| "eval_runtime": 10.4668, | |
| "eval_samples_per_second": 23.407, | |
| "eval_steps_per_second": 2.962, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27852998065764023, | |
| "grad_norm": 5.231250286102295, | |
| "learning_rate": 0.00025135126257861296, | |
| "loss": 7.7243, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28884590586718245, | |
| "grad_norm": 4.705758094787598, | |
| "learning_rate": 0.00024762976829079836, | |
| "loss": 7.6729, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28884590586718245, | |
| "eval_loss": 7.607565402984619, | |
| "eval_runtime": 96.7081, | |
| "eval_samples_per_second": 2.533, | |
| "eval_steps_per_second": 0.321, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29916183107672467, | |
| "grad_norm": 5.811936855316162, | |
| "learning_rate": 0.00024380095323321433, | |
| "loss": 7.6601, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "grad_norm": 6.612433433532715, | |
| "learning_rate": 0.00023986902627971652, | |
| "loss": 7.7225, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30947775628626695, | |
| "eval_loss": 7.7170515060424805, | |
| "eval_runtime": 102.1676, | |
| "eval_samples_per_second": 2.398, | |
| "eval_steps_per_second": 0.303, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31979368149580917, | |
| "grad_norm": 22.143421173095703, | |
| "learning_rate": 0.00023583830965122902, | |
| "loss": 7.7705, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3301096067053514, | |
| "grad_norm": 14.632792472839355, | |
| "learning_rate": 0.000231713234164488, | |
| "loss": 7.8259, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3301096067053514, | |
| "eval_loss": 7.772371292114258, | |
| "eval_runtime": 92.9761, | |
| "eval_samples_per_second": 2.635, | |
| "eval_steps_per_second": 0.333, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 12.084867477416992, | |
| "learning_rate": 0.00022749833436140976, | |
| "loss": 7.753, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.35074145712443583, | |
| "grad_norm": 9.957462310791016, | |
| "learning_rate": 0.00022319824352443768, | |
| "loss": 7.6611, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.35074145712443583, | |
| "eval_loss": 7.5973920822143555, | |
| "eval_runtime": 108.0726, | |
| "eval_samples_per_second": 2.267, | |
| "eval_steps_per_second": 0.287, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3610573823339781, | |
| "grad_norm": 4.924723148345947, | |
| "learning_rate": 0.0002188176885833471, | |
| "loss": 7.6173, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3713733075435203, | |
| "grad_norm": 9.477831840515137, | |
| "learning_rate": 0.0002143614849191077, | |
| "loss": 7.5696, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3713733075435203, | |
| "eval_loss": 7.603182792663574, | |
| "eval_runtime": 91.6779, | |
| "eval_samples_per_second": 2.672, | |
| "eval_steps_per_second": 0.338, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.38168923275306255, | |
| "grad_norm": 7.13740348815918, | |
| "learning_rate": 0.00020983453107051425, | |
| "loss": 7.6439, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39200515796260477, | |
| "grad_norm": 10.324462890625, | |
| "learning_rate": 0.00020524180334940528, | |
| "loss": 7.6786, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39200515796260477, | |
| "eval_loss": 7.616292476654053, | |
| "eval_runtime": 106.0137, | |
| "eval_samples_per_second": 2.311, | |
| "eval_steps_per_second": 0.292, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.402321083172147, | |
| "grad_norm": 6.7738938331604, | |
| "learning_rate": 0.00020058835037038873, | |
| "loss": 7.6126, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4126370083816892, | |
| "grad_norm": 4.637758255004883, | |
| "learning_rate": 0.00019587928750108816, | |
| "loss": 7.4746, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4126370083816892, | |
| "eval_loss": 7.426816463470459, | |
| "eval_runtime": 104.9303, | |
| "eval_samples_per_second": 2.335, | |
| "eval_steps_per_second": 0.295, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4229529335912315, | |
| "grad_norm": 10.853010177612305, | |
| "learning_rate": 0.000191119791239009, | |
| "loss": 7.4631, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.4332688588007737, | |
| "grad_norm": 17.25995635986328, | |
| "learning_rate": 0.000186315093521208, | |
| "loss": 7.4383, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4332688588007737, | |
| "eval_loss": 7.406857013702393, | |
| "eval_runtime": 10.9396, | |
| "eval_samples_per_second": 22.396, | |
| "eval_steps_per_second": 2.834, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4435847840103159, | |
| "grad_norm": 48.24932098388672, | |
| "learning_rate": 0.00018147047597301952, | |
| "loss": 7.5079, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45390070921985815, | |
| "grad_norm": 10.03960132598877, | |
| "learning_rate": 0.00017659126410216118, | |
| "loss": 7.4469, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45390070921985815, | |
| "eval_loss": 7.522530555725098, | |
| "eval_runtime": 11.0052, | |
| "eval_samples_per_second": 22.262, | |
| "eval_steps_per_second": 2.817, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46421663442940037, | |
| "grad_norm": 37.15389633178711, | |
| "learning_rate": 0.00017168282144460167, | |
| "loss": 7.4621, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47453255963894264, | |
| "grad_norm": 17.32857894897461, | |
| "learning_rate": 0.00016675054366862551, | |
| "loss": 7.6465, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47453255963894264, | |
| "eval_loss": 7.484851837158203, | |
| "eval_runtime": 11.0169, | |
| "eval_samples_per_second": 22.239, | |
| "eval_steps_per_second": 2.814, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 8.746517181396484, | |
| "learning_rate": 0.0001617998526435754, | |
| "loss": 7.5273, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4951644100580271, | |
| "grad_norm": 10.883841514587402, | |
| "learning_rate": 0.00015683619047979322, | |
| "loss": 7.4025, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4951644100580271, | |
| "eval_loss": 7.30989933013916, | |
| "eval_runtime": 10.9893, | |
| "eval_samples_per_second": 22.294, | |
| "eval_steps_per_second": 2.821, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5054803352675693, | |
| "grad_norm": 33.5476188659668, | |
| "learning_rate": 0.000151865013546311, | |
| "loss": 7.4748, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5157962604771116, | |
| "grad_norm": 9.489357948303223, | |
| "learning_rate": 0.00014689178647286702, | |
| "loss": 7.3473, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5157962604771116, | |
| "eval_loss": 7.262345314025879, | |
| "eval_runtime": 11.0031, | |
| "eval_samples_per_second": 22.267, | |
| "eval_steps_per_second": 2.817, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5261121856866537, | |
| "grad_norm": 17.06060791015625, | |
| "learning_rate": 0.00014192197614284246, | |
| "loss": 7.3059, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.536428110896196, | |
| "grad_norm": 6.841980934143066, | |
| "learning_rate": 0.00013696104568371937, | |
| "loss": 7.2821, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.536428110896196, | |
| "eval_loss": 7.248399257659912, | |
| "eval_runtime": 10.9335, | |
| "eval_samples_per_second": 22.408, | |
| "eval_steps_per_second": 2.835, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5467440361057382, | |
| "grad_norm": 30.315967559814453, | |
| "learning_rate": 0.00013201444846166842, | |
| "loss": 7.2134, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5570599613152805, | |
| "grad_norm": 58.75193786621094, | |
| "learning_rate": 0.00012708762208686638, | |
| "loss": 7.389, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5570599613152805, | |
| "eval_loss": 7.7176690101623535, | |
| "eval_runtime": 11.0125, | |
| "eval_samples_per_second": 22.247, | |
| "eval_steps_per_second": 2.815, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5673758865248227, | |
| "grad_norm": 12.3959379196167, | |
| "learning_rate": 0.00012218598243613358, | |
| "loss": 7.7131, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5776918117343649, | |
| "grad_norm": 6.3016510009765625, | |
| "learning_rate": 0.00011731491769946225, | |
| "loss": 7.2912, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5776918117343649, | |
| "eval_loss": 7.1140642166137695, | |
| "eval_runtime": 10.9151, | |
| "eval_samples_per_second": 22.446, | |
| "eval_steps_per_second": 2.84, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5880077369439072, | |
| "grad_norm": 12.621047019958496, | |
| "learning_rate": 0.00011247978245697986, | |
| "loss": 7.1747, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5983236621534493, | |
| "grad_norm": 11.38538646697998, | |
| "learning_rate": 0.00010768589179285843, | |
| "loss": 7.1847, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5983236621534493, | |
| "eval_loss": 7.114475250244141, | |
| "eval_runtime": 10.877, | |
| "eval_samples_per_second": 22.525, | |
| "eval_steps_per_second": 2.85, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6086395873629916, | |
| "grad_norm": 14.617621421813965, | |
| "learning_rate": 0.0001029385154526404, | |
| "loss": 7.1773, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "grad_norm": 8.394628524780273, | |
| "learning_rate": 9.824287205040372e-05, | |
| "loss": 7.2121, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6189555125725339, | |
| "eval_loss": 7.146487236022949, | |
| "eval_runtime": 10.8983, | |
| "eval_samples_per_second": 22.481, | |
| "eval_steps_per_second": 2.844, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6292714377820761, | |
| "grad_norm": 9.852251052856445, | |
| "learning_rate": 9.360412333213324e-05, | |
| "loss": 7.2931, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6395873629916183, | |
| "grad_norm": 9.686322212219238, | |
| "learning_rate": 8.902736850160597e-05, | |
| "loss": 7.1216, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6395873629916183, | |
| "eval_loss": 7.147933006286621, | |
| "eval_runtime": 10.9167, | |
| "eval_samples_per_second": 22.443, | |
| "eval_steps_per_second": 2.84, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6499032882011605, | |
| "grad_norm": 13.205273628234863, | |
| "learning_rate": 8.451763861502603e-05, | |
| "loss": 7.4023, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6602192134107028, | |
| "grad_norm": 7.406120300292969, | |
| "learning_rate": 8.007989105057155e-05, | |
| "loss": 7.2503, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6602192134107028, | |
| "eval_loss": 7.110508441925049, | |
| "eval_runtime": 10.8603, | |
| "eval_samples_per_second": 22.559, | |
| "eval_steps_per_second": 2.854, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.670535138620245, | |
| "grad_norm": 7.229670524597168, | |
| "learning_rate": 7.571900405893403e-05, | |
| "loss": 7.2731, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 13.080286979675293, | |
| "learning_rate": 7.143977140083847e-05, | |
| "loss": 7.1416, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "eval_loss": 7.173031806945801, | |
| "eval_runtime": 10.8292, | |
| "eval_samples_per_second": 22.624, | |
| "eval_steps_per_second": 2.863, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6911669890393295, | |
| "grad_norm": 7.729543685913086, | |
| "learning_rate": 6.724689707744056e-05, | |
| "loss": 7.1927, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7014829142488717, | |
| "grad_norm": 6.9558587074279785, | |
| "learning_rate": 6.314499015939392e-05, | |
| "loss": 7.2288, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7014829142488717, | |
| "eval_loss": 7.149050235748291, | |
| "eval_runtime": 101.2615, | |
| "eval_samples_per_second": 2.419, | |
| "eval_steps_per_second": 0.306, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7117988394584139, | |
| "grad_norm": 37.099365234375, | |
| "learning_rate": 5.913855972026981e-05, | |
| "loss": 7.3082, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7221147646679562, | |
| "grad_norm": 20.466567993164062, | |
| "learning_rate": 5.52320098799004e-05, | |
| "loss": 7.3502, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7221147646679562, | |
| "eval_loss": 7.199110984802246, | |
| "eval_runtime": 101.1627, | |
| "eval_samples_per_second": 2.422, | |
| "eval_steps_per_second": 0.306, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7324306898774984, | |
| "grad_norm": 9.72053337097168, | |
| "learning_rate": 5.1429634963094275e-05, | |
| "loss": 7.1518, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7427466150870407, | |
| "grad_norm": 8.377337455749512, | |
| "learning_rate": 4.773561477904451e-05, | |
| "loss": 7.2648, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7427466150870407, | |
| "eval_loss": 7.140583038330078, | |
| "eval_runtime": 102.1903, | |
| "eval_samples_per_second": 2.397, | |
| "eval_steps_per_second": 0.303, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7530625402965828, | |
| "grad_norm": 11.30972671508789, | |
| "learning_rate": 4.415401002662016e-05, | |
| "loss": 7.2351, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7633784655061251, | |
| "grad_norm": 11.135313987731934, | |
| "learning_rate": 4.068875783059153e-05, | |
| "loss": 7.1647, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7633784655061251, | |
| "eval_loss": 7.122550964355469, | |
| "eval_runtime": 94.8476, | |
| "eval_samples_per_second": 2.583, | |
| "eval_steps_per_second": 0.327, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 3.4541268348693848, | |
| "learning_rate": 3.734366741369488e-05, | |
| "loss": 7.0893, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7840103159252095, | |
| "grad_norm": 9.31264591217041, | |
| "learning_rate": 3.4122415909296155e-05, | |
| "loss": 7.1678, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7840103159252095, | |
| "eval_loss": 7.084330081939697, | |
| "eval_runtime": 97.4225, | |
| "eval_samples_per_second": 2.515, | |
| "eval_steps_per_second": 0.318, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7943262411347518, | |
| "grad_norm": 4.439526081085205, | |
| "learning_rate": 3.1028544319255097e-05, | |
| "loss": 7.1907, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.804642166344294, | |
| "grad_norm": 7.1955766677856445, | |
| "learning_rate": 2.8065453621433405e-05, | |
| "loss": 7.1879, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.804642166344294, | |
| "eval_loss": 7.1045451164245605, | |
| "eval_runtime": 98.3747, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 0.315, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8149580915538363, | |
| "grad_norm": 7.633655548095703, | |
| "learning_rate": 2.5236401031126713e-05, | |
| "loss": 7.1379, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8252740167633784, | |
| "grad_norm": 6.028517246246338, | |
| "learning_rate": 2.2544496420529107e-05, | |
| "loss": 7.2384, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8252740167633784, | |
| "eval_loss": 7.113685607910156, | |
| "eval_runtime": 104.0901, | |
| "eval_samples_per_second": 2.354, | |
| "eval_steps_per_second": 0.298, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8355899419729207, | |
| "grad_norm": 13.628016471862793, | |
| "learning_rate": 1.9992698900165983e-05, | |
| "loss": 7.0224, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.845905867182463, | |
| "grad_norm": 9.757091522216797, | |
| "learning_rate": 1.7583813566054454e-05, | |
| "loss": 7.2301, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.845905867182463, | |
| "eval_loss": 7.094879150390625, | |
| "eval_runtime": 11.0422, | |
| "eval_samples_per_second": 22.188, | |
| "eval_steps_per_second": 2.807, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8562217923920051, | |
| "grad_norm": 12.490655899047852, | |
| "learning_rate": 1.5320488416165604e-05, | |
| "loss": 7.2313, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8665377176015474, | |
| "grad_norm": 17.20357894897461, | |
| "learning_rate": 1.3205211439578656e-05, | |
| "loss": 7.2897, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8665377176015474, | |
| "eval_loss": 7.127319812774658, | |
| "eval_runtime": 11.0618, | |
| "eval_samples_per_second": 22.148, | |
| "eval_steps_per_second": 2.802, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8768536428110896, | |
| "grad_norm": 9.043377876281738, | |
| "learning_rate": 1.1240307881527516e-05, | |
| "loss": 7.0591, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8871695680206318, | |
| "grad_norm": 6.038938522338867, | |
| "learning_rate": 9.427937687345144e-06, | |
| "loss": 7.1483, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8871695680206318, | |
| "eval_loss": 7.108406066894531, | |
| "eval_runtime": 11.0489, | |
| "eval_samples_per_second": 22.174, | |
| "eval_steps_per_second": 2.806, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8974854932301741, | |
| "grad_norm": 6.139174461364746, | |
| "learning_rate": 7.770093128115911e-06, | |
| "loss": 7.1885, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9078014184397163, | |
| "grad_norm": 9.17184829711914, | |
| "learning_rate": 6.268596610646382e-06, | |
| "loss": 7.1119, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9078014184397163, | |
| "eval_loss": 7.098378658294678, | |
| "eval_runtime": 11.0353, | |
| "eval_samples_per_second": 22.201, | |
| "eval_steps_per_second": 2.809, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9181173436492586, | |
| "grad_norm": 9.084051132202148, | |
| "learning_rate": 4.9250986741612985e-06, | |
| "loss": 7.0772, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "grad_norm": 4.8953070640563965, | |
| "learning_rate": 3.7410761759270934e-06, | |
| "loss": 7.2202, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9284332688588007, | |
| "eval_loss": 7.076570510864258, | |
| "eval_runtime": 11.0328, | |
| "eval_samples_per_second": 22.206, | |
| "eval_steps_per_second": 2.81, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.938749194068343, | |
| "grad_norm": 3.9921817779541016, | |
| "learning_rate": 2.7178306677978767e-06, | |
| "loss": 7.1844, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9490651192778853, | |
| "grad_norm": 6.253467559814453, | |
| "learning_rate": 1.8564869654679181e-06, | |
| "loss": 7.1149, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9490651192778853, | |
| "eval_loss": 7.073785781860352, | |
| "eval_runtime": 10.9871, | |
| "eval_samples_per_second": 22.299, | |
| "eval_steps_per_second": 2.821, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9593810444874274, | |
| "grad_norm": 5.679617881774902, | |
| "learning_rate": 1.157991912003453e-06, | |
| "loss": 7.204, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "grad_norm": 10.825676918029785, | |
| "learning_rate": 6.231133370135422e-07, | |
| "loss": 7.1986, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "eval_loss": 7.074854850769043, | |
| "eval_runtime": 11.0251, | |
| "eval_samples_per_second": 22.222, | |
| "eval_steps_per_second": 2.812, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9800128949065119, | |
| "grad_norm": 3.1997151374816895, | |
| "learning_rate": 2.524392126034891e-07, | |
| "loss": 7.289, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9903288201160542, | |
| "grad_norm": 5.230665683746338, | |
| "learning_rate": 4.63770070389724e-08, | |
| "loss": 7.155, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9903288201160542, | |
| "eval_loss": 7.072939872741699, | |
| "eval_runtime": 11.0357, | |
| "eval_samples_per_second": 22.201, | |
| "eval_steps_per_second": 2.809, | |
| "step": 384 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 387, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.288233513590456e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |