|
{ |
|
"best_metric": 4.9799017906188965, |
|
"best_model_checkpoint": "/home/datta0/models/lora_final/gemma-2-9b_pct_default_r32/checkpoint-8", |
|
"epoch": 0.9992756539235412, |
|
"eval_steps": 8, |
|
"global_step": 388, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.002575452716297787, |
|
"grad_norm": 10.302180290222168, |
|
"learning_rate": 1.25e-05, |
|
"loss": 2.1766, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.010301810865191148, |
|
"grad_norm": 7.941323280334473, |
|
"learning_rate": 5e-05, |
|
"loss": 2.2338, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.020603621730382295, |
|
"grad_norm": 11.684494018554688, |
|
"learning_rate": 0.0001, |
|
"loss": 2.5978, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.020603621730382295, |
|
"eval_loss": 4.9799017906188965, |
|
"eval_runtime": 409.6588, |
|
"eval_samples_per_second": 0.598, |
|
"eval_steps_per_second": 0.598, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.03090543259557344, |
|
"grad_norm": 11.442267417907715, |
|
"learning_rate": 9.997266286704631e-05, |
|
"loss": 7.9986, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 0.04120724346076459, |
|
"grad_norm": 1.5122904777526855, |
|
"learning_rate": 9.989068136093873e-05, |
|
"loss": 11.0056, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 0.04120724346076459, |
|
"eval_loss": 11.464006423950195, |
|
"eval_runtime": 170.0933, |
|
"eval_samples_per_second": 1.44, |
|
"eval_steps_per_second": 1.44, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 0.05150905432595573, |
|
"grad_norm": 1.6047507524490356, |
|
"learning_rate": 9.975414512725057e-05, |
|
"loss": 11.4322, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.06181086519114688, |
|
"grad_norm": 1.1092612743377686, |
|
"learning_rate": 9.956320346634876e-05, |
|
"loss": 11.7215, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 0.06181086519114688, |
|
"eval_loss": 11.87649917602539, |
|
"eval_runtime": 121.7891, |
|
"eval_samples_per_second": 2.012, |
|
"eval_steps_per_second": 2.012, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 0.07211267605633803, |
|
"grad_norm": 0.7658454179763794, |
|
"learning_rate": 9.931806517013612e-05, |
|
"loss": 11.7902, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.08241448692152918, |
|
"grad_norm": 0.6134170889854431, |
|
"learning_rate": 9.901899829374047e-05, |
|
"loss": 11.8793, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 0.08241448692152918, |
|
"eval_loss": 11.905890464782715, |
|
"eval_runtime": 167.278, |
|
"eval_samples_per_second": 1.465, |
|
"eval_steps_per_second": 1.465, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 0.09271629778672032, |
|
"grad_norm": 0.7564122080802917, |
|
"learning_rate": 9.86663298624003e-05, |
|
"loss": 11.9183, |
|
"step": 36 |
|
}, |
|
{ |
|
"epoch": 0.10301810865191147, |
|
"grad_norm": 0.5781177282333374, |
|
"learning_rate": 9.826044551386744e-05, |
|
"loss": 11.8739, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.10301810865191147, |
|
"eval_loss": 11.97806167602539, |
|
"eval_runtime": 123.4702, |
|
"eval_samples_per_second": 1.984, |
|
"eval_steps_per_second": 1.984, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.11331991951710262, |
|
"grad_norm": 0.399111807346344, |
|
"learning_rate": 9.780178907671789e-05, |
|
"loss": 11.9112, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 0.12362173038229377, |
|
"grad_norm": 0.5412446856498718, |
|
"learning_rate": 9.729086208503174e-05, |
|
"loss": 11.8763, |
|
"step": 48 |
|
}, |
|
{ |
|
"epoch": 0.12362173038229377, |
|
"eval_loss": 11.9486665725708, |
|
"eval_runtime": 166.3239, |
|
"eval_samples_per_second": 1.473, |
|
"eval_steps_per_second": 1.473, |
|
"step": 48 |
|
}, |
|
{ |
|
"epoch": 0.1339235412474849, |
|
"grad_norm": 0.324489563703537, |
|
"learning_rate": 9.672822322997305e-05, |
|
"loss": 11.8606, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.14422535211267606, |
|
"grad_norm": 0.22089843451976776, |
|
"learning_rate": 9.611448774886924e-05, |
|
"loss": 11.8231, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.14422535211267606, |
|
"eval_loss": 11.828161239624023, |
|
"eval_runtime": 119.3481, |
|
"eval_samples_per_second": 2.053, |
|
"eval_steps_per_second": 2.053, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.1545271629778672, |
|
"grad_norm": 0.22084373235702515, |
|
"learning_rate": 9.545032675245813e-05, |
|
"loss": 11.7885, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.16482897384305836, |
|
"grad_norm": 0.5548378229141235, |
|
"learning_rate": 9.473646649103818e-05, |
|
"loss": 11.7758, |
|
"step": 64 |
|
}, |
|
{ |
|
"epoch": 0.16482897384305836, |
|
"eval_loss": 11.76635456085205, |
|
"eval_runtime": 123.4461, |
|
"eval_samples_per_second": 1.985, |
|
"eval_steps_per_second": 1.985, |
|
"step": 64 |
|
}, |
|
{ |
|
"epoch": 0.1751307847082495, |
|
"grad_norm": 0.22932168841362, |
|
"learning_rate": 9.397368756032445e-05, |
|
"loss": 11.7211, |
|
"step": 68 |
|
}, |
|
{ |
|
"epoch": 0.18543259557344063, |
|
"grad_norm": 0.33755362033843994, |
|
"learning_rate": 9.316282404787871e-05, |
|
"loss": 11.8011, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 0.18543259557344063, |
|
"eval_loss": 11.437198638916016, |
|
"eval_runtime": 167.588, |
|
"eval_samples_per_second": 1.462, |
|
"eval_steps_per_second": 1.462, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 0.1957344064386318, |
|
"grad_norm": 0.2544913589954376, |
|
"learning_rate": 9.230476262104677e-05, |
|
"loss": 11.7226, |
|
"step": 76 |
|
}, |
|
{ |
|
"epoch": 0.20603621730382293, |
|
"grad_norm": 0.2166949361562729, |
|
"learning_rate": 9.140044155740101e-05, |
|
"loss": 11.6991, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.20603621730382293, |
|
"eval_loss": 11.733440399169922, |
|
"eval_runtime": 121.1371, |
|
"eval_samples_per_second": 2.023, |
|
"eval_steps_per_second": 2.023, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.2163380281690141, |
|
"grad_norm": 0.1701814830303192, |
|
"learning_rate": 9.045084971874738e-05, |
|
"loss": 11.6466, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 0.22663983903420523, |
|
"grad_norm": 0.2655963599681854, |
|
"learning_rate": 8.945702546981969e-05, |
|
"loss": 11.8108, |
|
"step": 88 |
|
}, |
|
{ |
|
"epoch": 0.22663983903420523, |
|
"eval_loss": 11.500480651855469, |
|
"eval_runtime": 169.1725, |
|
"eval_samples_per_second": 1.448, |
|
"eval_steps_per_second": 1.448, |
|
"step": 88 |
|
}, |
|
{ |
|
"epoch": 0.23694164989939637, |
|
"grad_norm": 0.162321999669075, |
|
"learning_rate": 8.842005554284296e-05, |
|
"loss": 11.6505, |
|
"step": 92 |
|
}, |
|
{ |
|
"epoch": 0.24724346076458753, |
|
"grad_norm": 0.21160480380058289, |
|
"learning_rate": 8.73410738492077e-05, |
|
"loss": 11.6519, |
|
"step": 96 |
|
}, |
|
{ |
|
"epoch": 0.24724346076458753, |
|
"eval_loss": 11.594391822814941, |
|
"eval_runtime": 123.5651, |
|
"eval_samples_per_second": 1.983, |
|
"eval_steps_per_second": 1.983, |
|
"step": 96 |
|
}, |
|
{ |
|
"epoch": 0.2575452716297787, |
|
"grad_norm": 0.2764425277709961, |
|
"learning_rate": 8.622126023955446e-05, |
|
"loss": 11.6293, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.2678470824949698, |
|
"grad_norm": 0.179239422082901, |
|
"learning_rate": 8.506183921362443e-05, |
|
"loss": 11.6905, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 0.2678470824949698, |
|
"eval_loss": 11.594428062438965, |
|
"eval_runtime": 166.9483, |
|
"eval_samples_per_second": 1.468, |
|
"eval_steps_per_second": 1.468, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 0.27814889336016096, |
|
"grad_norm": 0.2203957438468933, |
|
"learning_rate": 8.386407858128706e-05, |
|
"loss": 11.564, |
|
"step": 108 |
|
}, |
|
{ |
|
"epoch": 0.28845070422535213, |
|
"grad_norm": 0.1474369317293167, |
|
"learning_rate": 8.262928807620843e-05, |
|
"loss": 11.6003, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 0.28845070422535213, |
|
"eval_loss": 11.591402053833008, |
|
"eval_runtime": 125.6068, |
|
"eval_samples_per_second": 1.951, |
|
"eval_steps_per_second": 1.951, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 0.29875251509054324, |
|
"grad_norm": 0.18921791017055511, |
|
"learning_rate": 8.135881792367686e-05, |
|
"loss": 11.5716, |
|
"step": 116 |
|
}, |
|
{ |
|
"epoch": 0.3090543259557344, |
|
"grad_norm": 0.1806280016899109, |
|
"learning_rate": 8.005405736415126e-05, |
|
"loss": 11.5813, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.3090543259557344, |
|
"eval_loss": 11.568363189697266, |
|
"eval_runtime": 122.1447, |
|
"eval_samples_per_second": 2.006, |
|
"eval_steps_per_second": 2.006, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.31935613682092556, |
|
"grad_norm": 0.1429259330034256, |
|
"learning_rate": 7.871643313414718e-05, |
|
"loss": 11.5332, |
|
"step": 124 |
|
}, |
|
{ |
|
"epoch": 0.3296579476861167, |
|
"grad_norm": 0.34623807668685913, |
|
"learning_rate": 7.734740790612136e-05, |
|
"loss": 11.5493, |
|
"step": 128 |
|
}, |
|
{ |
|
"epoch": 0.3296579476861167, |
|
"eval_loss": 11.755960464477539, |
|
"eval_runtime": 168.7681, |
|
"eval_samples_per_second": 1.452, |
|
"eval_steps_per_second": 1.452, |
|
"step": 128 |
|
}, |
|
{ |
|
"epoch": 0.33995975855130783, |
|
"grad_norm": 0.12925852835178375, |
|
"learning_rate": 7.594847868906076e-05, |
|
"loss": 11.6452, |
|
"step": 132 |
|
}, |
|
{ |
|
"epoch": 0.350261569416499, |
|
"grad_norm": 0.2058892697095871, |
|
"learning_rate": 7.452117519152542e-05, |
|
"loss": 11.5458, |
|
"step": 136 |
|
}, |
|
{ |
|
"epoch": 0.350261569416499, |
|
"eval_loss": 11.456594467163086, |
|
"eval_runtime": 125.1196, |
|
"eval_samples_per_second": 1.958, |
|
"eval_steps_per_second": 1.958, |
|
"step": 136 |
|
}, |
|
{ |
|
"epoch": 0.36056338028169016, |
|
"grad_norm": 0.1822740137577057, |
|
"learning_rate": 7.30670581489344e-05, |
|
"loss": 11.549, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.37086519114688127, |
|
"grad_norm": 0.163137748837471, |
|
"learning_rate": 7.158771761692464e-05, |
|
"loss": 11.5838, |
|
"step": 144 |
|
}, |
|
{ |
|
"epoch": 0.37086519114688127, |
|
"eval_loss": 11.433055877685547, |
|
"eval_runtime": 168.7124, |
|
"eval_samples_per_second": 1.452, |
|
"eval_steps_per_second": 1.452, |
|
"step": 144 |
|
}, |
|
{ |
|
"epoch": 0.38116700201207243, |
|
"grad_norm": 0.15767963230609894, |
|
"learning_rate": 7.008477123264848e-05, |
|
"loss": 11.5418, |
|
"step": 148 |
|
}, |
|
{ |
|
"epoch": 0.3914688128772636, |
|
"grad_norm": 0.14546583592891693, |
|
"learning_rate": 6.855986244591104e-05, |
|
"loss": 11.4815, |
|
"step": 152 |
|
}, |
|
{ |
|
"epoch": 0.3914688128772636, |
|
"eval_loss": 11.51737117767334, |
|
"eval_runtime": 123.7345, |
|
"eval_samples_per_second": 1.98, |
|
"eval_steps_per_second": 1.98, |
|
"step": 152 |
|
}, |
|
{ |
|
"epoch": 0.4017706237424547, |
|
"grad_norm": 0.1699499934911728, |
|
"learning_rate": 6.701465872208216e-05, |
|
"loss": 11.5552, |
|
"step": 156 |
|
}, |
|
{ |
|
"epoch": 0.41207243460764587, |
|
"grad_norm": 0.1260180026292801, |
|
"learning_rate": 6.545084971874738e-05, |
|
"loss": 11.5369, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.41207243460764587, |
|
"eval_loss": 11.527098655700684, |
|
"eval_runtime": 171.5714, |
|
"eval_samples_per_second": 1.428, |
|
"eval_steps_per_second": 1.428, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.42237424547283703, |
|
"grad_norm": 0.15433725714683533, |
|
"learning_rate": 6.387014543809223e-05, |
|
"loss": 11.4842, |
|
"step": 164 |
|
}, |
|
{ |
|
"epoch": 0.4326760563380282, |
|
"grad_norm": 0.12766706943511963, |
|
"learning_rate": 6.227427435703997e-05, |
|
"loss": 11.4617, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 0.4326760563380282, |
|
"eval_loss": 11.539175987243652, |
|
"eval_runtime": 123.9703, |
|
"eval_samples_per_second": 1.976, |
|
"eval_steps_per_second": 1.976, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 0.4429778672032193, |
|
"grad_norm": 0.16595180332660675, |
|
"learning_rate": 6.066498153718735e-05, |
|
"loss": 11.5383, |
|
"step": 172 |
|
}, |
|
{ |
|
"epoch": 0.45327967806841046, |
|
"grad_norm": 0.25058212876319885, |
|
"learning_rate": 5.90440267166055e-05, |
|
"loss": 11.4399, |
|
"step": 176 |
|
}, |
|
{ |
|
"epoch": 0.45327967806841046, |
|
"eval_loss": 11.36911678314209, |
|
"eval_runtime": 123.5694, |
|
"eval_samples_per_second": 1.983, |
|
"eval_steps_per_second": 1.983, |
|
"step": 176 |
|
}, |
|
{ |
|
"epoch": 0.4635814889336016, |
|
"grad_norm": 0.15986469388008118, |
|
"learning_rate": 5.74131823855921e-05, |
|
"loss": 11.4953, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.47388329979879273, |
|
"grad_norm": 20.536422729492188, |
|
"learning_rate": 5.577423184847932e-05, |
|
"loss": 11.3199, |
|
"step": 184 |
|
}, |
|
{ |
|
"epoch": 0.47388329979879273, |
|
"eval_loss": 10.98233413696289, |
|
"eval_runtime": 171.9363, |
|
"eval_samples_per_second": 1.425, |
|
"eval_steps_per_second": 1.425, |
|
"step": 184 |
|
}, |
|
{ |
|
"epoch": 0.4841851106639839, |
|
"grad_norm": 5.964752674102783, |
|
"learning_rate": 5.4128967273616625e-05, |
|
"loss": 10.8849, |
|
"step": 188 |
|
}, |
|
{ |
|
"epoch": 0.49448692152917506, |
|
"grad_norm": 6.385639190673828, |
|
"learning_rate": 5.247918773366112e-05, |
|
"loss": 10.6547, |
|
"step": 192 |
|
}, |
|
{ |
|
"epoch": 0.49448692152917506, |
|
"eval_loss": 10.066610336303711, |
|
"eval_runtime": 123.4211, |
|
"eval_samples_per_second": 1.985, |
|
"eval_steps_per_second": 1.985, |
|
"step": 192 |
|
}, |
|
{ |
|
"epoch": 0.5047887323943662, |
|
"grad_norm": 12.043314933776855, |
|
"learning_rate": 5.0826697238317935e-05, |
|
"loss": 9.5631, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 0.5150905432595574, |
|
"grad_norm": 7.537539482116699, |
|
"learning_rate": 4.917330276168208e-05, |
|
"loss": 8.8163, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.5150905432595574, |
|
"eval_loss": 8.763765335083008, |
|
"eval_runtime": 170.1114, |
|
"eval_samples_per_second": 1.44, |
|
"eval_steps_per_second": 1.44, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.5253923541247485, |
|
"grad_norm": 1.9154856204986572, |
|
"learning_rate": 4.7520812266338885e-05, |
|
"loss": 9.183, |
|
"step": 204 |
|
}, |
|
{ |
|
"epoch": 0.5356941649899396, |
|
"grad_norm": 4.3046746253967285, |
|
"learning_rate": 4.5871032726383386e-05, |
|
"loss": 9.5635, |
|
"step": 208 |
|
}, |
|
{ |
|
"epoch": 0.5356941649899396, |
|
"eval_loss": 8.515321731567383, |
|
"eval_runtime": 124.7005, |
|
"eval_samples_per_second": 1.965, |
|
"eval_steps_per_second": 1.965, |
|
"step": 208 |
|
}, |
|
{ |
|
"epoch": 0.5459959758551308, |
|
"grad_norm": 3.1673686504364014, |
|
"learning_rate": 4.4225768151520694e-05, |
|
"loss": 8.7257, |
|
"step": 212 |
|
}, |
|
{ |
|
"epoch": 0.5562977867203219, |
|
"grad_norm": 3.7206709384918213, |
|
"learning_rate": 4.2586817614407895e-05, |
|
"loss": 8.7862, |
|
"step": 216 |
|
}, |
|
{ |
|
"epoch": 0.5562977867203219, |
|
"eval_loss": 9.218582153320312, |
|
"eval_runtime": 170.5652, |
|
"eval_samples_per_second": 1.436, |
|
"eval_steps_per_second": 1.436, |
|
"step": 216 |
|
}, |
|
{ |
|
"epoch": 0.566599597585513, |
|
"grad_norm": 1.9827094078063965, |
|
"learning_rate": 4.095597328339452e-05, |
|
"loss": 9.8043, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.5769014084507043, |
|
"grad_norm": 3.734112501144409, |
|
"learning_rate": 3.933501846281267e-05, |
|
"loss": 10.2774, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 0.5769014084507043, |
|
"eval_loss": 10.383406639099121, |
|
"eval_runtime": 124.0491, |
|
"eval_samples_per_second": 1.975, |
|
"eval_steps_per_second": 1.975, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 0.5872032193158954, |
|
"grad_norm": 8.66118049621582, |
|
"learning_rate": 3.772572564296005e-05, |
|
"loss": 10.1576, |
|
"step": 228 |
|
}, |
|
{ |
|
"epoch": 0.5975050301810865, |
|
"grad_norm": 4.829358100891113, |
|
"learning_rate": 3.612985456190778e-05, |
|
"loss": 9.7932, |
|
"step": 232 |
|
}, |
|
{ |
|
"epoch": 0.5975050301810865, |
|
"eval_loss": 9.797175407409668, |
|
"eval_runtime": 124.9878, |
|
"eval_samples_per_second": 1.96, |
|
"eval_steps_per_second": 1.96, |
|
"step": 232 |
|
}, |
|
{ |
|
"epoch": 0.6078068410462777, |
|
"grad_norm": 2.6055920124053955, |
|
"learning_rate": 3.4549150281252636e-05, |
|
"loss": 9.7817, |
|
"step": 236 |
|
}, |
|
{ |
|
"epoch": 0.6181086519114688, |
|
"grad_norm": 5.202971458435059, |
|
"learning_rate": 3.298534127791785e-05, |
|
"loss": 9.5421, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.6181086519114688, |
|
"eval_loss": 9.684460639953613, |
|
"eval_runtime": 169.5135, |
|
"eval_samples_per_second": 1.445, |
|
"eval_steps_per_second": 1.445, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.6284104627766599, |
|
"grad_norm": 4.3098225593566895, |
|
"learning_rate": 3.144013755408895e-05, |
|
"loss": 9.574, |
|
"step": 244 |
|
}, |
|
{ |
|
"epoch": 0.6387122736418511, |
|
"grad_norm": 4.262834072113037, |
|
"learning_rate": 2.991522876735154e-05, |
|
"loss": 9.5401, |
|
"step": 248 |
|
}, |
|
{ |
|
"epoch": 0.6387122736418511, |
|
"eval_loss": 9.343782424926758, |
|
"eval_runtime": 123.888, |
|
"eval_samples_per_second": 1.978, |
|
"eval_steps_per_second": 1.978, |
|
"step": 248 |
|
}, |
|
{ |
|
"epoch": 0.6490140845070422, |
|
"grad_norm": 2.082047462463379, |
|
"learning_rate": 2.8412282383075363e-05, |
|
"loss": 9.9613, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 0.6593158953722335, |
|
"grad_norm": 0.8725143074989319, |
|
"learning_rate": 2.693294185106562e-05, |
|
"loss": 10.9001, |
|
"step": 256 |
|
}, |
|
{ |
|
"epoch": 0.6593158953722335, |
|
"eval_loss": 10.607041358947754, |
|
"eval_runtime": 171.7905, |
|
"eval_samples_per_second": 1.426, |
|
"eval_steps_per_second": 1.426, |
|
"step": 256 |
|
}, |
|
{ |
|
"epoch": 0.6696177062374246, |
|
"grad_norm": 5.7337517738342285, |
|
"learning_rate": 2.547882480847461e-05, |
|
"loss": 10.5149, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.6799195171026157, |
|
"grad_norm": 2.850618839263916, |
|
"learning_rate": 2.405152131093926e-05, |
|
"loss": 9.959, |
|
"step": 264 |
|
}, |
|
{ |
|
"epoch": 0.6799195171026157, |
|
"eval_loss": 9.617180824279785, |
|
"eval_runtime": 123.5003, |
|
"eval_samples_per_second": 1.984, |
|
"eval_steps_per_second": 1.984, |
|
"step": 264 |
|
}, |
|
{ |
|
"epoch": 0.6902213279678069, |
|
"grad_norm": 4.845343112945557, |
|
"learning_rate": 2.2652592093878666e-05, |
|
"loss": 9.2827, |
|
"step": 268 |
|
}, |
|
{ |
|
"epoch": 0.700523138832998, |
|
"grad_norm": 2.2851219177246094, |
|
"learning_rate": 2.128356686585282e-05, |
|
"loss": 9.5409, |
|
"step": 272 |
|
}, |
|
{ |
|
"epoch": 0.700523138832998, |
|
"eval_loss": 10.476217269897461, |
|
"eval_runtime": 171.3145, |
|
"eval_samples_per_second": 1.43, |
|
"eval_steps_per_second": 1.43, |
|
"step": 272 |
|
}, |
|
{ |
|
"epoch": 0.7108249496981891, |
|
"grad_norm": 1.766412615776062, |
|
"learning_rate": 1.9945942635848748e-05, |
|
"loss": 10.9182, |
|
"step": 276 |
|
}, |
|
{ |
|
"epoch": 0.7211267605633803, |
|
"grad_norm": 1.4967565536499023, |
|
"learning_rate": 1.8641182076323148e-05, |
|
"loss": 10.8074, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.7211267605633803, |
|
"eval_loss": 10.487205505371094, |
|
"eval_runtime": 125.9842, |
|
"eval_samples_per_second": 1.945, |
|
"eval_steps_per_second": 1.945, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.7314285714285714, |
|
"grad_norm": 6.666006088256836, |
|
"learning_rate": 1.7370711923791567e-05, |
|
"loss": 10.3232, |
|
"step": 284 |
|
}, |
|
{ |
|
"epoch": 0.7417303822937625, |
|
"grad_norm": 4.8097381591796875, |
|
"learning_rate": 1.6135921418712956e-05, |
|
"loss": 9.1645, |
|
"step": 288 |
|
}, |
|
{ |
|
"epoch": 0.7417303822937625, |
|
"eval_loss": 7.656736373901367, |
|
"eval_runtime": 124.6349, |
|
"eval_samples_per_second": 1.966, |
|
"eval_steps_per_second": 1.966, |
|
"step": 288 |
|
}, |
|
{ |
|
"epoch": 0.7520321931589538, |
|
"grad_norm": 5.842891693115234, |
|
"learning_rate": 1.4938160786375572e-05, |
|
"loss": 7.4315, |
|
"step": 292 |
|
}, |
|
{ |
|
"epoch": 0.7623340040241449, |
|
"grad_norm": 4.397885322570801, |
|
"learning_rate": 1.3778739760445552e-05, |
|
"loss": 8.1072, |
|
"step": 296 |
|
}, |
|
{ |
|
"epoch": 0.7623340040241449, |
|
"eval_loss": 8.791058540344238, |
|
"eval_runtime": 169.9555, |
|
"eval_samples_per_second": 1.442, |
|
"eval_steps_per_second": 1.442, |
|
"step": 296 |
|
}, |
|
{ |
|
"epoch": 0.772635814889336, |
|
"grad_norm": 2.3736143112182617, |
|
"learning_rate": 1.2658926150792322e-05, |
|
"loss": 9.0859, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.7829376257545272, |
|
"grad_norm": 3.1305336952209473, |
|
"learning_rate": 1.157994445715706e-05, |
|
"loss": 9.7069, |
|
"step": 304 |
|
}, |
|
{ |
|
"epoch": 0.7829376257545272, |
|
"eval_loss": 10.004293441772461, |
|
"eval_runtime": 124.8947, |
|
"eval_samples_per_second": 1.962, |
|
"eval_steps_per_second": 1.962, |
|
"step": 304 |
|
}, |
|
{ |
|
"epoch": 0.7932394366197183, |
|
"grad_norm": 1.8686987161636353, |
|
"learning_rate": 1.0542974530180327e-05, |
|
"loss": 10.0589, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 0.8035412474849094, |
|
"grad_norm": 2.3722739219665527, |
|
"learning_rate": 9.549150281252633e-06, |
|
"loss": 10.0752, |
|
"step": 312 |
|
}, |
|
{ |
|
"epoch": 0.8035412474849094, |
|
"eval_loss": 10.165539741516113, |
|
"eval_runtime": 169.0313, |
|
"eval_samples_per_second": 1.449, |
|
"eval_steps_per_second": 1.449, |
|
"step": 312 |
|
}, |
|
{ |
|
"epoch": 0.8138430583501006, |
|
"grad_norm": 2.35557222366333, |
|
"learning_rate": 8.599558442598998e-06, |
|
"loss": 10.2222, |
|
"step": 316 |
|
}, |
|
{ |
|
"epoch": 0.8241448692152917, |
|
"grad_norm": 3.20621395111084, |
|
"learning_rate": 7.695237378953223e-06, |
|
"loss": 9.9734, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.8241448692152917, |
|
"eval_loss": 9.964385032653809, |
|
"eval_runtime": 125.1152, |
|
"eval_samples_per_second": 1.958, |
|
"eval_steps_per_second": 1.958, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.834446680080483, |
|
"grad_norm": 4.180117130279541, |
|
"learning_rate": 6.837175952121306e-06, |
|
"loss": 10.0431, |
|
"step": 324 |
|
}, |
|
{ |
|
"epoch": 0.8447484909456741, |
|
"grad_norm": 3.4952783584594727, |
|
"learning_rate": 6.026312439675552e-06, |
|
"loss": 9.6722, |
|
"step": 328 |
|
}, |
|
{ |
|
"epoch": 0.8447484909456741, |
|
"eval_loss": 9.780319213867188, |
|
"eval_runtime": 171.5492, |
|
"eval_samples_per_second": 1.428, |
|
"eval_steps_per_second": 1.428, |
|
"step": 328 |
|
}, |
|
{ |
|
"epoch": 0.8550503018108652, |
|
"grad_norm": 3.512146472930908, |
|
"learning_rate": 5.263533508961827e-06, |
|
"loss": 9.7029, |
|
"step": 332 |
|
}, |
|
{ |
|
"epoch": 0.8653521126760564, |
|
"grad_norm": 2.8218512535095215, |
|
"learning_rate": 4.549673247541875e-06, |
|
"loss": 9.8279, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 0.8653521126760564, |
|
"eval_loss": 9.582645416259766, |
|
"eval_runtime": 121.2292, |
|
"eval_samples_per_second": 2.021, |
|
"eval_steps_per_second": 2.021, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 0.8756539235412475, |
|
"grad_norm": 7.174349784851074, |
|
"learning_rate": 3.885512251130763e-06, |
|
"loss": 9.5723, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.8859557344064386, |
|
"grad_norm": 2.031242847442627, |
|
"learning_rate": 3.271776770026963e-06, |
|
"loss": 9.6714, |
|
"step": 344 |
|
}, |
|
{ |
|
"epoch": 0.8859557344064386, |
|
"eval_loss": 9.553346633911133, |
|
"eval_runtime": 126.1884, |
|
"eval_samples_per_second": 1.942, |
|
"eval_steps_per_second": 1.942, |
|
"step": 344 |
|
}, |
|
{ |
|
"epoch": 0.8962575452716298, |
|
"grad_norm": 12.41804313659668, |
|
"learning_rate": 2.7091379149682685e-06, |
|
"loss": 9.6239, |
|
"step": 348 |
|
}, |
|
{ |
|
"epoch": 0.9065593561368209, |
|
"grad_norm": 5.308027744293213, |
|
"learning_rate": 2.1982109232821178e-06, |
|
"loss": 9.655, |
|
"step": 352 |
|
}, |
|
{ |
|
"epoch": 0.9065593561368209, |
|
"eval_loss": 9.633858680725098, |
|
"eval_runtime": 172.0221, |
|
"eval_samples_per_second": 1.424, |
|
"eval_steps_per_second": 1.424, |
|
"step": 352 |
|
}, |
|
{ |
|
"epoch": 0.916861167002012, |
|
"grad_norm": 3.773082971572876, |
|
"learning_rate": 1.7395544861325718e-06, |
|
"loss": 9.7297, |
|
"step": 356 |
|
}, |
|
{ |
|
"epoch": 0.9271629778672033, |
|
"grad_norm": 3.001129150390625, |
|
"learning_rate": 1.333670137599713e-06, |
|
"loss": 9.7184, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.9271629778672033, |
|
"eval_loss": 9.7650728225708, |
|
"eval_runtime": 120.2567, |
|
"eval_samples_per_second": 2.037, |
|
"eval_steps_per_second": 2.037, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.9374647887323944, |
|
"grad_norm": 4.141623497009277, |
|
"learning_rate": 9.810017062595322e-07, |
|
"loss": 9.9158, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.9477665995975855, |
|
"grad_norm": 2.350005626678467, |
|
"learning_rate": 6.819348298638839e-07, |
|
"loss": 9.6142, |
|
"step": 368 |
|
}, |
|
{ |
|
"epoch": 0.9477665995975855, |
|
"eval_loss": 9.853599548339844, |
|
"eval_runtime": 175.2573, |
|
"eval_samples_per_second": 1.398, |
|
"eval_steps_per_second": 1.398, |
|
"step": 368 |
|
}, |
|
{ |
|
"epoch": 0.9580684104627767, |
|
"grad_norm": 3.0426127910614014, |
|
"learning_rate": 4.367965336512403e-07, |
|
"loss": 9.7248, |
|
"step": 372 |
|
}, |
|
{ |
|
"epoch": 0.9683702213279678, |
|
"grad_norm": 3.860062599182129, |
|
"learning_rate": 2.458548727494292e-07, |
|
"loss": 9.9249, |
|
"step": 376 |
|
}, |
|
{ |
|
"epoch": 0.9683702213279678, |
|
"eval_loss": 9.89064884185791, |
|
"eval_runtime": 125.4726, |
|
"eval_samples_per_second": 1.953, |
|
"eval_steps_per_second": 1.953, |
|
"step": 376 |
|
}, |
|
{ |
|
"epoch": 0.9786720321931589, |
|
"grad_norm": 1.843009114265442, |
|
"learning_rate": 1.0931863906127327e-07, |
|
"loss": 9.7664, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.9889738430583501, |
|
"grad_norm": 5.112157344818115, |
|
"learning_rate": 2.7337132953697554e-08, |
|
"loss": 9.8654, |
|
"step": 384 |
|
}, |
|
{ |
|
"epoch": 0.9889738430583501, |
|
"eval_loss": 9.901961326599121, |
|
"eval_runtime": 166.3626, |
|
"eval_samples_per_second": 1.473, |
|
"eval_steps_per_second": 1.473, |
|
"step": 384 |
|
}, |
|
{ |
|
"epoch": 0.9992756539235412, |
|
"grad_norm": 3.1725776195526123, |
|
"learning_rate": 0.0, |
|
"loss": 9.8706, |
|
"step": 388 |
|
} |
|
], |
|
"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": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 5.577366487952179e+17, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|