| { | |
| "best_global_step": 2600, | |
| "best_metric": 1.8000476360321045, | |
| "best_model_checkpoint": "./sft_output\\checkpoint-2600", | |
| "epoch": 1.5625, | |
| "eval_steps": 100, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006009615384615385, | |
| "grad_norm": 18.5, | |
| "learning_rate": 3.9e-06, | |
| "loss": 4.727504730224609, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01201923076923077, | |
| "grad_norm": 16.0, | |
| "learning_rate": 8.233333333333333e-06, | |
| "loss": 4.481142044067383, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018028846153846152, | |
| "grad_norm": 8.3125, | |
| "learning_rate": 1.2566666666666666e-05, | |
| "loss": 4.313880157470703, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02403846153846154, | |
| "grad_norm": 16.25, | |
| "learning_rate": 1.6899999999999997e-05, | |
| "loss": 4.401013946533203, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.030048076923076924, | |
| "grad_norm": 22.25, | |
| "learning_rate": 2.1233333333333332e-05, | |
| "loss": 3.5513683319091798, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.036057692307692304, | |
| "grad_norm": 9.5625, | |
| "learning_rate": 2.5566666666666664e-05, | |
| "loss": 3.0777938842773436, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.042067307692307696, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 2.99e-05, | |
| "loss": 3.072860908508301, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.04807692307692308, | |
| "grad_norm": 11.0625, | |
| "learning_rate": 3.423333333333333e-05, | |
| "loss": 2.960606002807617, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.054086538461538464, | |
| "grad_norm": 7.90625, | |
| "learning_rate": 3.8566666666666664e-05, | |
| "loss": 3.084041786193848, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06009615384615385, | |
| "grad_norm": 9.75, | |
| "learning_rate": 4.29e-05, | |
| "loss": 2.869779586791992, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06009615384615385, | |
| "eval_loss": 2.793410539627075, | |
| "eval_runtime": 3.715, | |
| "eval_samples_per_second": 398.112, | |
| "eval_steps_per_second": 49.798, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06610576923076923, | |
| "grad_norm": 9.6875, | |
| "learning_rate": 4.723333333333333e-05, | |
| "loss": 2.957443618774414, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.07211538461538461, | |
| "grad_norm": 9.25, | |
| "learning_rate": 5.156666666666666e-05, | |
| "loss": 2.967830276489258, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.078125, | |
| "grad_norm": 8.875, | |
| "learning_rate": 5.59e-05, | |
| "loss": 2.9085622787475587, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.08413461538461539, | |
| "grad_norm": 7.5625, | |
| "learning_rate": 6.0233333333333325e-05, | |
| "loss": 2.5667842864990233, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.09014423076923077, | |
| "grad_norm": 8.75, | |
| "learning_rate": 6.456666666666666e-05, | |
| "loss": 2.593577003479004, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.09615384615384616, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 6.4998713743806e-05, | |
| "loss": 2.601177787780762, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.10216346153846154, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 6.499426755681021e-05, | |
| "loss": 2.5299556732177733, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.10817307692307693, | |
| "grad_norm": 8.5625, | |
| "learning_rate": 6.498664599368945e-05, | |
| "loss": 2.628713607788086, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.1141826923076923, | |
| "grad_norm": 6.0, | |
| "learning_rate": 6.497584979923138e-05, | |
| "loss": 2.393807220458984, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.1201923076923077, | |
| "grad_norm": 16.75, | |
| "learning_rate": 6.496188002845219e-05, | |
| "loss": 2.466072845458984, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1201923076923077, | |
| "eval_loss": 2.491903305053711, | |
| "eval_runtime": 3.9407, | |
| "eval_samples_per_second": 375.313, | |
| "eval_steps_per_second": 46.946, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12620192307692307, | |
| "grad_norm": 9.1875, | |
| "learning_rate": 6.494473804649346e-05, | |
| "loss": 2.5577186584472655, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.13221153846153846, | |
| "grad_norm": 10.1875, | |
| "learning_rate": 6.492442552848885e-05, | |
| "loss": 2.6166301727294923, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.13822115384615385, | |
| "grad_norm": 7.96875, | |
| "learning_rate": 6.490094445940027e-05, | |
| "loss": 2.5525186538696287, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.14423076923076922, | |
| "grad_norm": 9.125, | |
| "learning_rate": 6.487429713382407e-05, | |
| "loss": 2.413387680053711, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1502403846153846, | |
| "grad_norm": 10.1875, | |
| "learning_rate": 6.484448615576662e-05, | |
| "loss": 2.5147281646728517, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 5.40625, | |
| "learning_rate": 6.481151443839005e-05, | |
| "loss": 2.433956336975098, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1622596153846154, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 6.477538520372738e-05, | |
| "loss": 2.53863525390625, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.16826923076923078, | |
| "grad_norm": 7.96875, | |
| "learning_rate": 6.47361019823678e-05, | |
| "loss": 2.4263484954833983, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.17427884615384615, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 6.469366861311157e-05, | |
| "loss": 2.4903234481811523, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.18028846153846154, | |
| "grad_norm": 6.78125, | |
| "learning_rate": 6.464808924259493e-05, | |
| "loss": 2.533559036254883, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.18028846153846154, | |
| "eval_loss": 2.3458189964294434, | |
| "eval_runtime": 4.0438, | |
| "eval_samples_per_second": 365.748, | |
| "eval_steps_per_second": 45.749, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.18629807692307693, | |
| "grad_norm": 8.875, | |
| "learning_rate": 6.459936832488491e-05, | |
| "loss": 2.5885854721069337, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 7.65625, | |
| "learning_rate": 6.4547510621044e-05, | |
| "loss": 2.290690231323242, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.19831730769230768, | |
| "grad_norm": 7.34375, | |
| "learning_rate": 6.449252119866495e-05, | |
| "loss": 2.4482185363769533, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.20432692307692307, | |
| "grad_norm": 7.34375, | |
| "learning_rate": 6.443440543137554e-05, | |
| "loss": 2.3000059127807617, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.21033653846153846, | |
| "grad_norm": 6.8125, | |
| "learning_rate": 6.437316899831347e-05, | |
| "loss": 2.4301517486572264, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.21634615384615385, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 6.43088178835714e-05, | |
| "loss": 2.4185550689697264, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.22235576923076922, | |
| "grad_norm": 6.375, | |
| "learning_rate": 6.424135837561213e-05, | |
| "loss": 2.3065673828125, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.2283653846153846, | |
| "grad_norm": 6.6875, | |
| "learning_rate": 6.417079706665414e-05, | |
| "loss": 2.4958147048950194, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.234375, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 6.40971408520274e-05, | |
| "loss": 2.2613336563110353, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 6.402039692949943e-05, | |
| "loss": 2.330600929260254, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "eval_loss": 2.247739791870117, | |
| "eval_runtime": 3.6579, | |
| "eval_samples_per_second": 404.336, | |
| "eval_steps_per_second": 50.576, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24639423076923078, | |
| "grad_norm": 8.5, | |
| "learning_rate": 6.39405727985721e-05, | |
| "loss": 2.443510055541992, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.25240384615384615, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 6.385767625974865e-05, | |
| "loss": 2.3513521194458007, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.25841346153846156, | |
| "grad_norm": 10.0625, | |
| "learning_rate": 6.377171541377146e-05, | |
| "loss": 2.023931694030762, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.2644230769230769, | |
| "grad_norm": 9.0, | |
| "learning_rate": 6.368269866083043e-05, | |
| "loss": 2.458607864379883, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.2704326923076923, | |
| "grad_norm": 8.875, | |
| "learning_rate": 6.35906346997421e-05, | |
| "loss": 2.2841283798217775, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.2764423076923077, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 6.349553252709957e-05, | |
| "loss": 2.1192956924438477, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.2824519230769231, | |
| "grad_norm": 7.875, | |
| "learning_rate": 6.339740143639341e-05, | |
| "loss": 2.362535095214844, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.28846153846153844, | |
| "grad_norm": 9.3125, | |
| "learning_rate": 6.329625101710341e-05, | |
| "loss": 2.456166076660156, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.29447115384615385, | |
| "grad_norm": 6.875, | |
| "learning_rate": 6.319209115376157e-05, | |
| "loss": 2.3226381301879884, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.3004807692307692, | |
| "grad_norm": 6.96875, | |
| "learning_rate": 6.308493202498606e-05, | |
| "loss": 2.2645195007324217, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3004807692307692, | |
| "eval_loss": 2.1779282093048096, | |
| "eval_runtime": 3.729, | |
| "eval_samples_per_second": 396.618, | |
| "eval_steps_per_second": 49.611, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.30649038461538464, | |
| "grad_norm": 6.8125, | |
| "learning_rate": 6.297478410248666e-05, | |
| "loss": 2.1719266891479494, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 6.625, | |
| "learning_rate": 6.28616581500414e-05, | |
| "loss": 2.289084625244141, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.31850961538461536, | |
| "grad_norm": 6.8125, | |
| "learning_rate": 6.274556522244473e-05, | |
| "loss": 2.1797677993774416, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.3245192307692308, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 6.262651666442723e-05, | |
| "loss": 2.21547908782959, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.33052884615384615, | |
| "grad_norm": 7.6875, | |
| "learning_rate": 6.250452410954698e-05, | |
| "loss": 1.92919979095459, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.33653846153846156, | |
| "grad_norm": 9.0, | |
| "learning_rate": 6.237959947905271e-05, | |
| "loss": 2.3097171783447266, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.3425480769230769, | |
| "grad_norm": 6.09375, | |
| "learning_rate": 6.225175498071885e-05, | |
| "loss": 2.0136302947998046, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.3485576923076923, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 6.212100310765263e-05, | |
| "loss": 2.0454858779907226, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.3545673076923077, | |
| "grad_norm": 8.375, | |
| "learning_rate": 6.198735663707312e-05, | |
| "loss": 2.1544803619384765, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.3605769230769231, | |
| "grad_norm": 9.0, | |
| "learning_rate": 6.185082862906272e-05, | |
| "loss": 2.2414445877075195, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3605769230769231, | |
| "eval_loss": 2.1135077476501465, | |
| "eval_runtime": 4.1967, | |
| "eval_samples_per_second": 352.42, | |
| "eval_steps_per_second": 44.082, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.36658653846153844, | |
| "grad_norm": 7.8125, | |
| "learning_rate": 6.171143242529091e-05, | |
| "loss": 2.204708480834961, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.37259615384615385, | |
| "grad_norm": 6.1875, | |
| "learning_rate": 6.156918164771042e-05, | |
| "loss": 2.0722976684570313, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.3786057692307692, | |
| "grad_norm": 8.25, | |
| "learning_rate": 6.142409019722617e-05, | |
| "loss": 2.019124412536621, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 6.1875, | |
| "learning_rate": 6.127617225233674e-05, | |
| "loss": 2.5064016342163087, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.390625, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 6.112544226774899e-05, | |
| "loss": 2.1757251739501955, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.39663461538461536, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 6.0971914972965364e-05, | |
| "loss": 2.228921127319336, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.4026442307692308, | |
| "grad_norm": 6.9375, | |
| "learning_rate": 6.081560537084464e-05, | |
| "loss": 2.1757333755493162, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.40865384615384615, | |
| "grad_norm": 6.46875, | |
| "learning_rate": 6.0656528736135715e-05, | |
| "loss": 1.9958133697509766, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.41466346153846156, | |
| "grad_norm": 7.15625, | |
| "learning_rate": 6.049470061398506e-05, | |
| "loss": 2.188085746765137, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.4206730769230769, | |
| "grad_norm": 7.59375, | |
| "learning_rate": 6.033013681841756e-05, | |
| "loss": 2.2390575408935547, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4206730769230769, | |
| "eval_loss": 2.0713417530059814, | |
| "eval_runtime": 4.1604, | |
| "eval_samples_per_second": 355.496, | |
| "eval_steps_per_second": 44.467, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4266826923076923, | |
| "grad_norm": 7.34375, | |
| "learning_rate": 6.016285343079112e-05, | |
| "loss": 2.424024772644043, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.4326923076923077, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 5.9992866798225276e-05, | |
| "loss": 2.0728567123413084, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.4387019230769231, | |
| "grad_norm": 7.875, | |
| "learning_rate": 5.982019353200365e-05, | |
| "loss": 1.842981719970703, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.44471153846153844, | |
| "grad_norm": 9.9375, | |
| "learning_rate": 5.964485050595071e-05, | |
| "loss": 2.1463977813720705, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.45072115384615385, | |
| "grad_norm": 6.46875, | |
| "learning_rate": 5.946685485478282e-05, | |
| "loss": 2.0530189514160155, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.4567307692307692, | |
| "grad_norm": 6.90625, | |
| "learning_rate": 5.9286223972433876e-05, | |
| "loss": 1.996546173095703, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.46274038461538464, | |
| "grad_norm": 6.90625, | |
| "learning_rate": 5.910297551035548e-05, | |
| "loss": 2.183040428161621, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 5.8917127375792035e-05, | |
| "loss": 2.1800765991210938, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.47475961538461536, | |
| "grad_norm": 8.0, | |
| "learning_rate": 5.872869773003088e-05, | |
| "loss": 2.0184604644775392, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 5.853770498662749e-05, | |
| "loss": 2.2337005615234373, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "eval_loss": 2.028367280960083, | |
| "eval_runtime": 3.5366, | |
| "eval_samples_per_second": 418.203, | |
| "eval_steps_per_second": 52.311, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.48677884615384615, | |
| "grad_norm": 6.59375, | |
| "learning_rate": 5.83441678096061e-05, | |
| "loss": 1.8032636642456055, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.49278846153846156, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 5.814810511163585e-05, | |
| "loss": 2.1184009552001952, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.4987980769230769, | |
| "grad_norm": 7.4375, | |
| "learning_rate": 5.794953605218262e-05, | |
| "loss": 1.8946741104125977, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.5048076923076923, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 5.77484800356367e-05, | |
| "loss": 1.8021522521972657, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.5108173076923077, | |
| "grad_norm": 8.125, | |
| "learning_rate": 5.754495670941665e-05, | |
| "loss": 2.1481500625610352, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.5168269230769231, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 5.733898596204924e-05, | |
| "loss": 1.9111835479736328, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.5228365384615384, | |
| "grad_norm": 8.0, | |
| "learning_rate": 5.7130587921226016e-05, | |
| "loss": 2.175729179382324, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.5288461538461539, | |
| "grad_norm": 6.8125, | |
| "learning_rate": 5.6919782951836314e-05, | |
| "loss": 2.1691364288330077, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.5348557692307693, | |
| "grad_norm": 8.125, | |
| "learning_rate": 5.6706591653977224e-05, | |
| "loss": 1.8160417556762696, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.5408653846153846, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 5.649103486094055e-05, | |
| "loss": 2.0434226989746094, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5408653846153846, | |
| "eval_loss": 2.005472421646118, | |
| "eval_runtime": 3.5088, | |
| "eval_samples_per_second": 421.515, | |
| "eval_steps_per_second": 52.725, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.546875, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 5.6273133637176894e-05, | |
| "loss": 2.134675407409668, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.5528846153846154, | |
| "grad_norm": 6.09375, | |
| "learning_rate": 5.605290927623727e-05, | |
| "loss": 1.733789825439453, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.5588942307692307, | |
| "grad_norm": 7.75, | |
| "learning_rate": 5.583038329869223e-05, | |
| "loss": 2.0644378662109375, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.5649038461538461, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 5.560557745002889e-05, | |
| "loss": 1.9948148727416992, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.5709134615384616, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 5.53785136985259e-05, | |
| "loss": 2.0190046310424803, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 5.514921423310672e-05, | |
| "loss": 1.9832630157470703, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.5829326923076923, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 5.491770146117127e-05, | |
| "loss": 1.9667003631591797, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.5889423076923077, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 5.468399800640622e-05, | |
| "loss": 1.983225440979004, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.5949519230769231, | |
| "grad_norm": 10.0625, | |
| "learning_rate": 5.444812670657427e-05, | |
| "loss": 2.0743431091308593, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.6009615384615384, | |
| "grad_norm": 7.59375, | |
| "learning_rate": 5.421011061128231e-05, | |
| "loss": 2.047312927246094, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6009615384615384, | |
| "eval_loss": 1.9628348350524902, | |
| "eval_runtime": 3.1807, | |
| "eval_samples_per_second": 464.991, | |
| "eval_steps_per_second": 58.163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6069711538461539, | |
| "grad_norm": 7.0625, | |
| "learning_rate": 5.396997297972907e-05, | |
| "loss": 1.8052534103393554, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.6129807692307693, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 5.372773727843215e-05, | |
| "loss": 2.065348434448242, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.6189903846153846, | |
| "grad_norm": 6.75, | |
| "learning_rate": 5.3483427178934894e-05, | |
| "loss": 2.2212120056152345, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 8.375, | |
| "learning_rate": 5.3237066555493144e-05, | |
| "loss": 1.900156593322754, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.6310096153846154, | |
| "grad_norm": 7.75, | |
| "learning_rate": 5.298867948274222e-05, | |
| "loss": 1.9202171325683595, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.6370192307692307, | |
| "grad_norm": 7.5625, | |
| "learning_rate": 5.273829023334437e-05, | |
| "loss": 2.100513458251953, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.6430288461538461, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 5.248592327561673e-05, | |
| "loss": 1.7703487396240234, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.6490384615384616, | |
| "grad_norm": 10.8125, | |
| "learning_rate": 5.223160327114033e-05, | |
| "loss": 1.8859819412231444, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.6550480769230769, | |
| "grad_norm": 7.9375, | |
| "learning_rate": 5.1975355072350126e-05, | |
| "loss": 2.0359310150146483, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.6610576923076923, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 5.171720372010633e-05, | |
| "loss": 1.9784713745117188, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6610576923076923, | |
| "eval_loss": 1.9280022382736206, | |
| "eval_runtime": 3.121, | |
| "eval_samples_per_second": 473.892, | |
| "eval_steps_per_second": 59.277, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6670673076923077, | |
| "grad_norm": 5.875, | |
| "learning_rate": 5.145717444124749e-05, | |
| "loss": 1.8144060134887696, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.6730769230769231, | |
| "grad_norm": 6.59375, | |
| "learning_rate": 5.11952926461252e-05, | |
| "loss": 2.0068008422851564, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.6790865384615384, | |
| "grad_norm": 7.90625, | |
| "learning_rate": 5.093158392612105e-05, | |
| "loss": 1.9808029174804687, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.6850961538461539, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 5.066607405114571e-05, | |
| "loss": 1.7743522644042968, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.6911057692307693, | |
| "grad_norm": 7.0, | |
| "learning_rate": 5.039878896712078e-05, | |
| "loss": 1.9338529586791993, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.6971153846153846, | |
| "grad_norm": 6.25, | |
| "learning_rate": 5.0129754793443277e-05, | |
| "loss": 1.9099365234375, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.703125, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 4.9858997820433155e-05, | |
| "loss": 2.0609180450439455, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.7091346153846154, | |
| "grad_norm": 7.5, | |
| "learning_rate": 4.9586544506764306e-05, | |
| "loss": 1.9219863891601563, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.7151442307692307, | |
| "grad_norm": 7.625, | |
| "learning_rate": 4.931242147687889e-05, | |
| "loss": 2.1477108001708984, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "grad_norm": 5.90625, | |
| "learning_rate": 4.903665551838561e-05, | |
| "loss": 1.979792022705078, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "eval_loss": 1.9053001403808594, | |
| "eval_runtime": 3.3287, | |
| "eval_samples_per_second": 444.314, | |
| "eval_steps_per_second": 55.577, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7271634615384616, | |
| "grad_norm": 6.3125, | |
| "learning_rate": 4.8759273579441983e-05, | |
| "loss": 1.6205293655395507, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.7331730769230769, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 4.848030276612095e-05, | |
| "loss": 1.877166748046875, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.7391826923076923, | |
| "grad_norm": 8.25, | |
| "learning_rate": 4.819977033976203e-05, | |
| "loss": 1.6161108016967773, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.7451923076923077, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 4.79177037143073e-05, | |
| "loss": 1.9132957458496094, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.7512019230769231, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 4.763413045362249e-05, | |
| "loss": 2.014938545227051, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.7572115384615384, | |
| "grad_norm": 5.9375, | |
| "learning_rate": 4.734907826880345e-05, | |
| "loss": 2.038423538208008, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.7632211538461539, | |
| "grad_norm": 6.9375, | |
| "learning_rate": 4.706257501546808e-05, | |
| "loss": 1.9666574478149415, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 4.677464869103435e-05, | |
| "loss": 2.2751462936401365, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.7752403846153846, | |
| "grad_norm": 5.78125, | |
| "learning_rate": 4.648532743198434e-05, | |
| "loss": 1.975306510925293, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 8.9375, | |
| "learning_rate": 4.619463951111467e-05, | |
| "loss": 1.8744926452636719, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "eval_loss": 1.8853334188461304, | |
| "eval_runtime": 3.3511, | |
| "eval_samples_per_second": 441.35, | |
| "eval_steps_per_second": 55.206, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7872596153846154, | |
| "grad_norm": 9.0, | |
| "learning_rate": 4.5902613334773705e-05, | |
| "loss": 2.0330549240112306, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.7932692307692307, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 4.5609277440085605e-05, | |
| "loss": 2.0024389266967773, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.7992788461538461, | |
| "grad_norm": 7.65625, | |
| "learning_rate": 4.53146604921617e-05, | |
| "loss": 1.763961410522461, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.8052884615384616, | |
| "grad_norm": 7.46875, | |
| "learning_rate": 4.501879128129926e-05, | |
| "loss": 1.7687673568725586, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.8112980769230769, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 4.472169872016808e-05, | |
| "loss": 2.0385513305664062, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.8173076923076923, | |
| "grad_norm": 9.125, | |
| "learning_rate": 4.4423411840985114e-05, | |
| "loss": 1.8624149322509767, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.8233173076923077, | |
| "grad_norm": 7.6875, | |
| "learning_rate": 4.412395979267741e-05, | |
| "loss": 1.9508615493774415, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.8293269230769231, | |
| "grad_norm": 6.6875, | |
| "learning_rate": 4.382337183803365e-05, | |
| "loss": 1.884683609008789, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.8353365384615384, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 4.352167735084454e-05, | |
| "loss": 1.7874622344970703, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.8413461538461539, | |
| "grad_norm": 7.15625, | |
| "learning_rate": 4.3218905813032395e-05, | |
| "loss": 2.23499755859375, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8413461538461539, | |
| "eval_loss": 1.8636605739593506, | |
| "eval_runtime": 3.4388, | |
| "eval_samples_per_second": 430.093, | |
| "eval_steps_per_second": 53.798, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8473557692307693, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 4.291508681177012e-05, | |
| "loss": 1.7971698760986328, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.8533653846153846, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 4.2610250036589923e-05, | |
| "loss": 1.9670312881469727, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.859375, | |
| "grad_norm": 8.0, | |
| "learning_rate": 4.2304425276481985e-05, | |
| "loss": 2.1206539154052733, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.8653846153846154, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 4.199764241698352e-05, | |
| "loss": 1.892685317993164, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.8713942307692307, | |
| "grad_norm": 9.6875, | |
| "learning_rate": 4.168993143725826e-05, | |
| "loss": 2.0682640075683594, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.8774038461538461, | |
| "grad_norm": 6.9375, | |
| "learning_rate": 4.1381322407166884e-05, | |
| "loss": 1.641856575012207, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.8834134615384616, | |
| "grad_norm": 7.46875, | |
| "learning_rate": 4.107184548432855e-05, | |
| "loss": 1.7966501235961914, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.8894230769230769, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 4.0761530911173895e-05, | |
| "loss": 1.928413200378418, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.8954326923076923, | |
| "grad_norm": 7.65625, | |
| "learning_rate": 4.045040901198965e-05, | |
| "loss": 1.8558208465576171, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.9014423076923077, | |
| "grad_norm": 7.8125, | |
| "learning_rate": 4.013851018995535e-05, | |
| "loss": 1.8771005630493165, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9014423076923077, | |
| "eval_loss": 1.8482130765914917, | |
| "eval_runtime": 4.1673, | |
| "eval_samples_per_second": 354.91, | |
| "eval_steps_per_second": 44.394, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9074519230769231, | |
| "grad_norm": 7.6875, | |
| "learning_rate": 3.9825864924172315e-05, | |
| "loss": 1.7130186080932617, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.9134615384615384, | |
| "grad_norm": 6.71875, | |
| "learning_rate": 3.951250376668518e-05, | |
| "loss": 1.6876375198364257, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.9194711538461539, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 3.919845733949629e-05, | |
| "loss": 1.9814535140991212, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.9254807692307693, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 3.888375633157335e-05, | |
| "loss": 1.6066390991210937, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.9314903846153846, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 3.856843149585039e-05, | |
| "loss": 1.8126174926757812, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 7.84375, | |
| "learning_rate": 3.825251364622262e-05, | |
| "loss": 1.9670969009399415, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.9435096153846154, | |
| "grad_norm": 8.0, | |
| "learning_rate": 3.7936033654535224e-05, | |
| "loss": 1.9797147750854491, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.9495192307692307, | |
| "grad_norm": 7.375, | |
| "learning_rate": 3.7619022447566504e-05, | |
| "loss": 1.8385818481445313, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.9555288461538461, | |
| "grad_norm": 9.1875, | |
| "learning_rate": 3.730151100400575e-05, | |
| "loss": 2.053624725341797, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 6.90625, | |
| "learning_rate": 3.6983530351425924e-05, | |
| "loss": 1.8292165756225587, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "eval_loss": 1.8361619710922241, | |
| "eval_runtime": 3.2911, | |
| "eval_samples_per_second": 449.398, | |
| "eval_steps_per_second": 56.213, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.9675480769230769, | |
| "grad_norm": 7.0, | |
| "learning_rate": 3.666511156325161e-05, | |
| "loss": 1.8061643600463868, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.9735576923076923, | |
| "grad_norm": 8.25, | |
| "learning_rate": 3.634628575572249e-05, | |
| "loss": 1.7992219924926758, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.9795673076923077, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 3.602708408485264e-05, | |
| "loss": 1.8782625198364258, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.9855769230769231, | |
| "grad_norm": 6.875, | |
| "learning_rate": 3.5707537743385914e-05, | |
| "loss": 1.9543542861938477, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.9915865384615384, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 3.538767795774776e-05, | |
| "loss": 1.8594795227050782, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.9975961538461539, | |
| "grad_norm": 8.5625, | |
| "learning_rate": 3.5067535984993755e-05, | |
| "loss": 1.914571762084961, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.0036057692307692, | |
| "grad_norm": 5.78125, | |
| "learning_rate": 3.474714310975511e-05, | |
| "loss": 1.6203495025634767, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.0096153846153846, | |
| "grad_norm": 5.53125, | |
| "learning_rate": 3.442653064118149e-05, | |
| "loss": 1.2878865242004394, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.015625, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 3.410572990988149e-05, | |
| "loss": 1.4174379348754882, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.0216346153846154, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 3.378477226486094e-05, | |
| "loss": 1.3761696815490723, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0216346153846154, | |
| "eval_loss": 1.8251831531524658, | |
| "eval_runtime": 3.9899, | |
| "eval_samples_per_second": 370.69, | |
| "eval_steps_per_second": 46.368, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0276442307692308, | |
| "grad_norm": 6.3125, | |
| "learning_rate": 3.3463689070459446e-05, | |
| "loss": 1.4362096786499023, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.0336538461538463, | |
| "grad_norm": 7.375, | |
| "learning_rate": 3.314251170328545e-05, | |
| "loss": 1.4190923690795898, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.0396634615384615, | |
| "grad_norm": 8.375, | |
| "learning_rate": 3.2821271549150065e-05, | |
| "loss": 1.491513442993164, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.0456730769230769, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 3.25e-05, | |
| "loss": 1.5360705375671386, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.0516826923076923, | |
| "grad_norm": 7.125, | |
| "learning_rate": 3.2178728450849936e-05, | |
| "loss": 1.4541635513305664, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.0576923076923077, | |
| "grad_norm": 7.96875, | |
| "learning_rate": 3.185748829671454e-05, | |
| "loss": 1.4496195793151856, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.0637019230769231, | |
| "grad_norm": 6.46875, | |
| "learning_rate": 3.153631092954055e-05, | |
| "loss": 1.3822254180908202, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.0697115384615385, | |
| "grad_norm": 6.75, | |
| "learning_rate": 3.1215227735139054e-05, | |
| "loss": 1.3735233306884767, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.0757211538461537, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 3.08942700901185e-05, | |
| "loss": 1.6167102813720704, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.0817307692307692, | |
| "grad_norm": 7.125, | |
| "learning_rate": 3.0573469358818515e-05, | |
| "loss": 1.4335213661193849, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.0817307692307692, | |
| "eval_loss": 1.8208625316619873, | |
| "eval_runtime": 4.1162, | |
| "eval_samples_per_second": 359.315, | |
| "eval_steps_per_second": 44.945, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.0877403846153846, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 3.0252856890244898e-05, | |
| "loss": 1.3433932304382323, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 6.75, | |
| "learning_rate": 2.9932464015006232e-05, | |
| "loss": 1.3711178779602051, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.0997596153846154, | |
| "grad_norm": 8.5, | |
| "learning_rate": 2.961232204225223e-05, | |
| "loss": 1.152962589263916, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.1057692307692308, | |
| "grad_norm": 9.5625, | |
| "learning_rate": 2.929246225661408e-05, | |
| "loss": 1.515265941619873, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.1117788461538463, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 2.8972915915147355e-05, | |
| "loss": 1.41017484664917, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.1177884615384615, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 2.8653714244277507e-05, | |
| "loss": 1.3034038543701172, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.1237980769230769, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 2.8334888436748388e-05, | |
| "loss": 1.524069881439209, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.1298076923076923, | |
| "grad_norm": 8.0625, | |
| "learning_rate": 2.8016469648574073e-05, | |
| "loss": 1.4052583694458007, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.1358173076923077, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 2.7698488995994248e-05, | |
| "loss": 1.5672178268432617, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.1418269230769231, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 2.7380977552433497e-05, | |
| "loss": 1.4612812042236327, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.1418269230769231, | |
| "eval_loss": 1.8157281875610352, | |
| "eval_runtime": 4.0121, | |
| "eval_samples_per_second": 368.633, | |
| "eval_steps_per_second": 46.11, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.1478365384615385, | |
| "grad_norm": 6.625, | |
| "learning_rate": 2.706396634546478e-05, | |
| "loss": 1.5970848083496094, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 2.6747486353777374e-05, | |
| "loss": 1.34104585647583, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.1598557692307692, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 2.6431568504149604e-05, | |
| "loss": 1.5024894714355468, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.1658653846153846, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 2.6116243668426652e-05, | |
| "loss": 1.4241928100585937, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.171875, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 2.5801542660503707e-05, | |
| "loss": 1.4265833854675294, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.1778846153846154, | |
| "grad_norm": 7.46875, | |
| "learning_rate": 2.5487496233314824e-05, | |
| "loss": 1.481613540649414, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.1838942307692308, | |
| "grad_norm": 7.90625, | |
| "learning_rate": 2.5174135075827676e-05, | |
| "loss": 1.4637072563171387, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.1899038461538463, | |
| "grad_norm": 6.78125, | |
| "learning_rate": 2.486148981004464e-05, | |
| "loss": 1.5539934158325195, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.1959134615384615, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 2.4549590988010345e-05, | |
| "loss": 1.4752711296081542, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.2019230769230769, | |
| "grad_norm": 8.375, | |
| "learning_rate": 2.42384690888261e-05, | |
| "loss": 1.4823905944824218, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2019230769230769, | |
| "eval_loss": 1.8108009099960327, | |
| "eval_runtime": 3.4473, | |
| "eval_samples_per_second": 429.035, | |
| "eval_steps_per_second": 53.666, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2079326923076923, | |
| "grad_norm": 7.90625, | |
| "learning_rate": 2.392815451567144e-05, | |
| "loss": 1.4651354789733886, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 1.2139423076923077, | |
| "grad_norm": 6.625, | |
| "learning_rate": 2.3618677592833117e-05, | |
| "loss": 1.4358355522155761, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.2199519230769231, | |
| "grad_norm": 5.375, | |
| "learning_rate": 2.3310068562741736e-05, | |
| "loss": 1.4444135665893554, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 1.2259615384615385, | |
| "grad_norm": 7.34375, | |
| "learning_rate": 2.3002357583016475e-05, | |
| "loss": 1.6786272048950195, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.2319711538461537, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 2.269557472351801e-05, | |
| "loss": 1.3511952400207519, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.2379807692307692, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 2.2389749963410078e-05, | |
| "loss": 1.246901035308838, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 1.2439903846153846, | |
| "grad_norm": 5.71875, | |
| "learning_rate": 2.2084913188229875e-05, | |
| "loss": 1.5045913696289062, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 2.1781094186967606e-05, | |
| "loss": 1.3058452606201172, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.2560096153846154, | |
| "grad_norm": 6.78125, | |
| "learning_rate": 2.1478322649155464e-05, | |
| "loss": 1.3314120292663574, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 1.2620192307692308, | |
| "grad_norm": 8.0, | |
| "learning_rate": 2.1176628161966357e-05, | |
| "loss": 1.5527116775512695, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.2620192307692308, | |
| "eval_loss": 1.8070952892303467, | |
| "eval_runtime": 3.1287, | |
| "eval_samples_per_second": 472.727, | |
| "eval_steps_per_second": 59.131, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.2680288461538463, | |
| "grad_norm": 5.625, | |
| "learning_rate": 2.087604020732259e-05, | |
| "loss": 1.305916976928711, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 1.2740384615384617, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 2.0576588159014884e-05, | |
| "loss": 1.4427724838256837, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.2800480769230769, | |
| "grad_norm": 7.25, | |
| "learning_rate": 2.027830127983191e-05, | |
| "loss": 1.4432353973388672, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 1.2860576923076923, | |
| "grad_norm": 6.65625, | |
| "learning_rate": 1.9981208718700723e-05, | |
| "loss": 1.5362314224243163, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 1.2920673076923077, | |
| "grad_norm": 6.25, | |
| "learning_rate": 1.9685339507838284e-05, | |
| "loss": 1.4252591133117676, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.2980769230769231, | |
| "grad_norm": 6.78125, | |
| "learning_rate": 1.9390722559914383e-05, | |
| "loss": 1.4497568130493164, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.3040865384615383, | |
| "grad_norm": 6.0, | |
| "learning_rate": 1.9097386665226293e-05, | |
| "loss": 1.2646799087524414, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 1.3100961538461537, | |
| "grad_norm": 5.875, | |
| "learning_rate": 1.880536048888534e-05, | |
| "loss": 1.3875950813293456, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 1.3161057692307692, | |
| "grad_norm": 5.28125, | |
| "learning_rate": 1.851467256801566e-05, | |
| "loss": 1.5071735382080078, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 1.3221153846153846, | |
| "grad_norm": 5.46875, | |
| "learning_rate": 1.822535130896565e-05, | |
| "loss": 1.331043815612793, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.3221153846153846, | |
| "eval_loss": 1.8051131963729858, | |
| "eval_runtime": 3.0379, | |
| "eval_samples_per_second": 486.849, | |
| "eval_steps_per_second": 60.897, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.328125, | |
| "grad_norm": 9.375, | |
| "learning_rate": 1.7937424984531922e-05, | |
| "loss": 1.4430349349975586, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 1.3341346153846154, | |
| "grad_norm": 5.75, | |
| "learning_rate": 1.7650921731196555e-05, | |
| "loss": 1.4894984245300293, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 1.3401442307692308, | |
| "grad_norm": 8.3125, | |
| "learning_rate": 1.7365869546377496e-05, | |
| "loss": 1.4858686447143554, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 1.7082296285692707e-05, | |
| "loss": 1.485416793823242, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 1.3521634615384617, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 1.6800229660237972e-05, | |
| "loss": 1.4124130249023437, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.3581730769230769, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 1.6519697233879057e-05, | |
| "loss": 1.3517161369323731, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 1.3641826923076923, | |
| "grad_norm": 13.4375, | |
| "learning_rate": 1.624072642055803e-05, | |
| "loss": 1.500981616973877, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 1.3701923076923077, | |
| "grad_norm": 6.5, | |
| "learning_rate": 1.5963344481614383e-05, | |
| "loss": 1.3381962776184082, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 1.3762019230769231, | |
| "grad_norm": 6.1875, | |
| "learning_rate": 1.5687578523121106e-05, | |
| "loss": 1.475235366821289, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 1.3822115384615383, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 1.541345549323568e-05, | |
| "loss": 1.4238420486450196, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.3822115384615383, | |
| "eval_loss": 1.8031173944473267, | |
| "eval_runtime": 3.9238, | |
| "eval_samples_per_second": 376.935, | |
| "eval_steps_per_second": 47.149, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.3882211538461537, | |
| "grad_norm": 6.59375, | |
| "learning_rate": 1.5141002179566836e-05, | |
| "loss": 1.2991218566894531, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 1.3942307692307692, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 1.4870245206556716e-05, | |
| "loss": 1.386824893951416, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 1.4002403846153846, | |
| "grad_norm": 5.78125, | |
| "learning_rate": 1.460121103287921e-05, | |
| "loss": 1.4255257606506349, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 7.40625, | |
| "learning_rate": 1.4333925948854295e-05, | |
| "loss": 1.4677438735961914, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 1.4122596153846154, | |
| "grad_norm": 5.625, | |
| "learning_rate": 1.4068416073878953e-05, | |
| "loss": 1.3717265129089355, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.4182692307692308, | |
| "grad_norm": 7.5, | |
| "learning_rate": 1.3804707353874798e-05, | |
| "loss": 1.2748692512512207, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 1.4242788461538463, | |
| "grad_norm": 6.5, | |
| "learning_rate": 1.3542825558752506e-05, | |
| "loss": 1.3941977500915528, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 1.4302884615384617, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 1.328279627989367e-05, | |
| "loss": 1.4263758659362793, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 1.4362980769230769, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 1.3024644927649877e-05, | |
| "loss": 1.4275604248046876, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "grad_norm": 7.53125, | |
| "learning_rate": 1.2768396728859671e-05, | |
| "loss": 1.5580693244934083, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "eval_loss": 1.8019680976867676, | |
| "eval_runtime": 3.2145, | |
| "eval_samples_per_second": 460.101, | |
| "eval_steps_per_second": 57.551, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4483173076923077, | |
| "grad_norm": 6.6875, | |
| "learning_rate": 1.251407672438327e-05, | |
| "loss": 1.3324359893798827, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 1.4543269230769231, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 1.2261709766655632e-05, | |
| "loss": 1.490361213684082, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 1.4603365384615383, | |
| "grad_norm": 6.09375, | |
| "learning_rate": 1.2011320517257769e-05, | |
| "loss": 1.4653941154479981, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 1.4663461538461537, | |
| "grad_norm": 7.0, | |
| "learning_rate": 1.1762933444506844e-05, | |
| "loss": 1.4430002212524413, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 1.4723557692307692, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 1.1516572821065102e-05, | |
| "loss": 1.3758784294128419, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.4783653846153846, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 1.1272262721567842e-05, | |
| "loss": 1.4369168281555176, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 1.484375, | |
| "grad_norm": 6.03125, | |
| "learning_rate": 1.1030027020270933e-05, | |
| "loss": 1.1694883346557616, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 1.4903846153846154, | |
| "grad_norm": 6.625, | |
| "learning_rate": 1.0789889388717683e-05, | |
| "loss": 1.3628769874572755, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 1.4963942307692308, | |
| "grad_norm": 5.9375, | |
| "learning_rate": 1.055187329342573e-05, | |
| "loss": 1.408741283416748, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 1.5024038461538463, | |
| "grad_norm": 6.625, | |
| "learning_rate": 1.0316001993593767e-05, | |
| "loss": 1.3211170196533204, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5024038461538463, | |
| "eval_loss": 1.8011430501937866, | |
| "eval_runtime": 4.0497, | |
| "eval_samples_per_second": 365.211, | |
| "eval_steps_per_second": 45.682, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5084134615384617, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 1.0082298538828732e-05, | |
| "loss": 1.495834732055664, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 1.5144230769230769, | |
| "grad_norm": 7.0625, | |
| "learning_rate": 9.85078576689328e-06, | |
| "loss": 1.4937422752380372, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.5204326923076923, | |
| "grad_norm": 6.625, | |
| "learning_rate": 9.621486301474094e-06, | |
| "loss": 1.363206672668457, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 1.5264423076923077, | |
| "grad_norm": 6.125, | |
| "learning_rate": 9.394422549971118e-06, | |
| "loss": 1.5537137985229492, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 1.5324519230769231, | |
| "grad_norm": 7.6875, | |
| "learning_rate": 9.169616701307767e-06, | |
| "loss": 1.440219783782959, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 7.65625, | |
| "learning_rate": 8.947090723762732e-06, | |
| "loss": 1.5468674659729005, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 1.5444711538461537, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 8.726866362823097e-06, | |
| "loss": 1.4238739013671875, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 1.5504807692307692, | |
| "grad_norm": 7.28125, | |
| "learning_rate": 8.508965139059449e-06, | |
| "loss": 1.4900738716125488, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 1.5564903846153846, | |
| "grad_norm": 7.0625, | |
| "learning_rate": 8.293408346022772e-06, | |
| "loss": 1.3469541549682618, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 15.375, | |
| "learning_rate": 8.080217048163685e-06, | |
| "loss": 1.44387845993042, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "eval_loss": 1.8000476360321045, | |
| "eval_runtime": 3.8414, | |
| "eval_samples_per_second": 385.019, | |
| "eval_steps_per_second": 48.16, | |
| "step": 2600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 3328, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 945013082701824.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |