| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 685, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0072992700729927005, | |
| "grad_norm": 708.0, | |
| "learning_rate": 2.898550724637681e-06, | |
| "loss": 56.8346, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0364963503649635, | |
| "grad_norm": 604.0, | |
| "learning_rate": 1.4492753623188407e-05, | |
| "loss": 52.9742, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.072992700729927, | |
| "grad_norm": 340.0, | |
| "learning_rate": 2.8985507246376814e-05, | |
| "loss": 39.0746, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10948905109489052, | |
| "grad_norm": 40.25, | |
| "learning_rate": 4.347826086956522e-05, | |
| "loss": 20.8099, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.145985401459854, | |
| "grad_norm": 25.5, | |
| "learning_rate": 5.797101449275363e-05, | |
| "loss": 17.6144, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18248175182481752, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 7.246376811594203e-05, | |
| "loss": 15.3803, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.21897810218978103, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 8.695652173913044e-05, | |
| "loss": 14.0798, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25547445255474455, | |
| "grad_norm": 13.4375, | |
| "learning_rate": 0.00010144927536231885, | |
| "loss": 13.4032, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.291970802919708, | |
| "grad_norm": 41.0, | |
| "learning_rate": 0.00011594202898550725, | |
| "loss": 10.8827, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3284671532846715, | |
| "grad_norm": 13.1875, | |
| "learning_rate": 0.00013043478260869567, | |
| "loss": 4.5915, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.36496350364963503, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 0.00014492753623188405, | |
| "loss": 1.9, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.40145985401459855, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 0.00015942028985507247, | |
| "loss": 1.6474, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.43795620437956206, | |
| "grad_norm": 3.5, | |
| "learning_rate": 0.00017391304347826088, | |
| "loss": 1.477, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4744525547445255, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 0.00018840579710144927, | |
| "loss": 1.3309, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5109489051094891, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 0.00019999869950890106, | |
| "loss": 1.2538, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5474452554744526, | |
| "grad_norm": 5.9375, | |
| "learning_rate": 0.0001999531858720213, | |
| "loss": 1.224, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.583941605839416, | |
| "grad_norm": 2.25, | |
| "learning_rate": 0.00019984268150178167, | |
| "loss": 1.1823, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6204379562043796, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 0.00019966725824941932, | |
| "loss": 1.1279, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.656934306569343, | |
| "grad_norm": 3.0625, | |
| "learning_rate": 0.00019942703017718975, | |
| "loss": 1.127, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6934306569343066, | |
| "grad_norm": 1.75, | |
| "learning_rate": 0.000199122153484202, | |
| "loss": 1.1284, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.7299270072992701, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 0.00019875282640485645, | |
| "loss": 1.0566, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7664233576642335, | |
| "grad_norm": 4.53125, | |
| "learning_rate": 0.0001983192890799503, | |
| "loss": 1.0361, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.8029197080291971, | |
| "grad_norm": 2.5, | |
| "learning_rate": 0.0001978218234005352, | |
| "loss": 1.0371, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8394160583941606, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 0.00019726075282462845, | |
| "loss": 1.0235, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8759124087591241, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 0.00019663644216689683, | |
| "loss": 0.996, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9124087591240876, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 0.00019594929736144976, | |
| "loss": 0.9734, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.948905109489051, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 0.00019519976519789616, | |
| "loss": 0.978, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9854014598540146, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 0.00019438833303083678, | |
| "loss": 0.9712, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.307734489440918, | |
| "eval_runtime": 0.9962, | |
| "eval_samples_per_second": 5.019, | |
| "eval_steps_per_second": 2.008, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 1.0218978102189782, | |
| "grad_norm": 2.125, | |
| "learning_rate": 0.00019351552846298025, | |
| "loss": 0.9374, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0583941605839415, | |
| "grad_norm": 2.265625, | |
| "learning_rate": 0.0001925819190020898, | |
| "loss": 0.9173, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.094890510948905, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00019158811169198313, | |
| "loss": 0.8916, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1313868613138687, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.0001905347527178252, | |
| "loss": 0.9418, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.167883211678832, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 0.00018942252698597113, | |
| "loss": 0.9054, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.2043795620437956, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 0.00018825215767863214, | |
| "loss": 0.9039, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.2408759124087592, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 0.00018702440578365387, | |
| "loss": 0.9146, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.2773722627737225, | |
| "grad_norm": 1.3515625, | |
| "learning_rate": 0.00018574006959971333, | |
| "loss": 0.8896, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.313868613138686, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 0.00018439998421725554, | |
| "loss": 0.8947, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.3503649635036497, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 0.00018300502097550806, | |
| "loss": 0.881, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.3868613138686132, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 0.00018155608689592604, | |
| "loss": 0.8906, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.4233576642335766, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 0.00018005412409243606, | |
| "loss": 0.8939, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.4598540145985401, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 0.0001785001091588628, | |
| "loss": 0.9016, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4963503649635037, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.0001768950525339362, | |
| "loss": 0.8943, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.5328467153284673, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 0.00017523999784429238, | |
| "loss": 0.8614, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.5693430656934306, | |
| "grad_norm": 0.7734375, | |
| "learning_rate": 0.00017353602122589527, | |
| "loss": 0.8788, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.6058394160583942, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.0001717842306243205, | |
| "loss": 0.8833, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.6423357664233578, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00016998576507435618, | |
| "loss": 0.8713, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.6788321167883211, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.00016814179395938913, | |
| "loss": 0.8661, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.7153284671532847, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 0.00016625351625105796, | |
| "loss": 0.8413, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.7518248175182483, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 0.0001643221597296679, | |
| "loss": 0.8741, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7883211678832116, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 0.00016234898018587337, | |
| "loss": 0.8744, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.8248175182481752, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 0.00016033526060414842, | |
| "loss": 0.8517, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8613138686131387, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 0.00015828231032857503, | |
| "loss": 0.8899, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.897810218978102, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 0.00015619146421149232, | |
| "loss": 0.8537, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.9343065693430657, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 0.00015406408174555976, | |
| "loss": 0.8329, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.9708029197080292, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 0.00015190154617979938, | |
| "loss": 0.8675, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.247941017150879, | |
| "eval_runtime": 0.9979, | |
| "eval_samples_per_second": 5.01, | |
| "eval_steps_per_second": 2.004, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 2.0072992700729926, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 0.00014970526362019079, | |
| "loss": 0.8435, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.0437956204379564, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 0.00014747666211540459, | |
| "loss": 0.7774, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0802919708029197, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 0.00014521719072826858, | |
| "loss": 0.79, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 2.116788321167883, | |
| "grad_norm": 0.498046875, | |
| "learning_rate": 0.00014292831859356997, | |
| "loss": 0.7929, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.153284671532847, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 0.00014061153396280674, | |
| "loss": 0.8032, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.18978102189781, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.7932, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.2262773722627736, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 0.00013590026998475986, | |
| "loss": 0.7657, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.2627737226277373, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0001335088539565523, | |
| "loss": 0.783, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.2992700729927007, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 0.00013109565007862596, | |
| "loss": 0.7755, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.335766423357664, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0001286622274444361, | |
| "loss": 0.7723, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.372262773722628, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 0.00012621016829391022, | |
| "loss": 0.7739, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.408759124087591, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 0.00012374106698465732, | |
| "loss": 0.7821, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.445255474452555, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 0.00012125652895529766, | |
| "loss": 0.7852, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.4817518248175183, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 0.00011875816968158815, | |
| "loss": 0.7792, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.5182481751824817, | |
| "grad_norm": 0.625, | |
| "learning_rate": 0.00011624761362602061, | |
| "loss": 0.7799, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.554744525547445, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 0.00011372649318157749, | |
| "loss": 0.7914, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.591240875912409, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 0.00011119644761033078, | |
| "loss": 0.7847, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.627737226277372, | |
| "grad_norm": 0.984375, | |
| "learning_rate": 0.0001086591219775746, | |
| "loss": 0.8049, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.664233576642336, | |
| "grad_norm": 0.81640625, | |
| "learning_rate": 0.00010611616608218429, | |
| "loss": 0.7865, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.7007299270072993, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00010356923338389806, | |
| "loss": 0.7908, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.7372262773722627, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 0.00010101997992821797, | |
| "loss": 0.7925, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.7737226277372264, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 9.847006326962974e-05, | |
| "loss": 0.799, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.81021897810219, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 9.592114139384145e-05, | |
| "loss": 0.7832, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.846715328467153, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 9.337487163974164e-05, | |
| "loss": 0.7796, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.883211678832117, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 9.083290962177828e-05, | |
| "loss": 0.7839, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.9197080291970803, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 8.829690815345886e-05, | |
| "loss": 0.7781, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.9562043795620436, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 8.57685161726715e-05, | |
| "loss": 0.7457, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.9927007299270074, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 8.324937766952638e-05, | |
| "loss": 0.7623, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.275648355484009, | |
| "eval_runtime": 0.9945, | |
| "eval_samples_per_second": 5.028, | |
| "eval_steps_per_second": 2.011, | |
| "step": 411 | |
| }, | |
| { | |
| "epoch": 3.0291970802919708, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 8.074113061741397e-05, | |
| "loss": 0.7329, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 3.065693430656934, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 7.824540590797568e-05, | |
| "loss": 0.7052, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.102189781021898, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 7.576382629067877e-05, | |
| "loss": 0.7015, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.1386861313868613, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 7.329800531768584e-05, | |
| "loss": 0.696, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.1751824817518246, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 7.084954629470417e-05, | |
| "loss": 0.7154, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 3.2116788321167884, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 6.842004123849752e-05, | |
| "loss": 0.7113, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.2481751824817517, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 6.601106984173835e-05, | |
| "loss": 0.7139, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 3.2846715328467155, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 6.362419844587287e-05, | |
| "loss": 0.6967, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.321167883211679, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 6.126097902266772e-05, | |
| "loss": 0.7073, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 3.3576642335766422, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 5.8922948165099524e-05, | |
| "loss": 0.6857, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.394160583941606, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 5.6611626088244194e-05, | |
| "loss": 0.7199, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 3.4306569343065694, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 5.432851564081534e-05, | |
| "loss": 0.7075, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.4671532846715327, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 5.207510132799436e-05, | |
| "loss": 0.7006, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.5036496350364965, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 4.9852848346187566e-05, | |
| "loss": 0.7151, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.54014598540146, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 4.7663201630338816e-05, | |
| "loss": 0.7129, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 3.576642335766423, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 4.550758491441526e-05, | |
| "loss": 0.7139, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.613138686131387, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 4.3387399805679255e-05, | |
| "loss": 0.7162, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 3.6496350364963503, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 4.1304024873346705e-05, | |
| "loss": 0.7132, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.686131386861314, | |
| "grad_norm": 0.57421875, | |
| "learning_rate": 3.9258814752225284e-05, | |
| "loss": 0.7007, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 3.7226277372262775, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 3.725309926191479e-05, | |
| "loss": 0.7037, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.759124087591241, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 3.528818254214329e-05, | |
| "loss": 0.7255, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 3.795620437956204, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 3.336534220479961e-05, | |
| "loss": 0.6966, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.832116788321168, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 3.1485828503215585e-05, | |
| "loss": 0.7143, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.8686131386861313, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 2.9650863519236418e-05, | |
| "loss": 0.7005, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.905109489051095, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 2.7861640368608844e-05, | |
| "loss": 0.7005, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 3.9416058394160585, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 2.6119322425203197e-05, | |
| "loss": 0.7139, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.978102189781022, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 2.4425042564574184e-05, | |
| "loss": 0.709, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.341665267944336, | |
| "eval_runtime": 0.9977, | |
| "eval_samples_per_second": 5.012, | |
| "eval_steps_per_second": 2.005, | |
| "step": 548 | |
| }, | |
| { | |
| "epoch": 4.014598540145985, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 2.277990242735185e-05, | |
| "loss": 0.6801, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.0510948905109485, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 2.118497170294195e-05, | |
| "loss": 0.6495, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 4.087591240875913, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 1.9641287434001355e-05, | |
| "loss": 0.672, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.124087591240876, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 1.8149853342140645e-05, | |
| "loss": 0.6611, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 4.160583941605839, | |
| "grad_norm": 0.59375, | |
| "learning_rate": 1.671163917529285e-05, | |
| "loss": 0.662, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.197080291970803, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 1.5327580077171587e-05, | |
| "loss": 0.6635, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 4.233576642335766, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 1.3998575979229944e-05, | |
| "loss": 0.6624, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.2700729927007295, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 1.272549101551438e-05, | |
| "loss": 0.6523, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 4.306569343065694, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 1.1509152960794666e-05, | |
| "loss": 0.6607, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.343065693430657, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 1.035035269233493e-05, | |
| "loss": 0.6626, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 4.37956204379562, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 9.249843675656212e-06, | |
| "loss": 0.678, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.416058394160584, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 8.208341474624071e-06, | |
| "loss": 0.6783, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 4.452554744525547, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 7.226523286180776e-06, | |
| "loss": 0.6699, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.489051094890511, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 6.3050275000238414e-06, | |
| "loss": 0.6607, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 4.525547445255475, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 5.4444532835175144e-06, | |
| "loss": 0.6702, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.562043795620438, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 4.6453601921072395e-06, | |
| "loss": 0.6793, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.598540145985401, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 3.908267805490051e-06, | |
| "loss": 0.6622, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.635036496350365, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 3.233655389777801e-06, | |
| "loss": 0.677, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 4.671532846715328, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 2.62196158587269e-06, | |
| "loss": 0.6588, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.708029197080292, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 2.073584124257899e-06, | |
| "loss": 0.6621, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 4.744525547445256, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 1.5888795663883904e-06, | |
| "loss": 0.6655, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.781021897810219, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 1.1681630728506699e-06, | |
| "loss": 0.6653, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 4.817518248175182, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 8.117081984415298e-07, | |
| "loss": 0.6734, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.854014598540146, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 5.19746714299596e-07, | |
| "loss": 0.6541, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 4.89051094890511, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 2.9246845720496407e-07, | |
| "loss": 0.6722, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.927007299270073, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 1.300212061451367e-07, | |
| "loss": 0.6472, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.963503649635037, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 3.251058622737446e-08, | |
| "loss": 0.667, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 0.0, | |
| "loss": 0.6601, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.3811252117156982, | |
| "eval_runtime": 0.9953, | |
| "eval_samples_per_second": 5.024, | |
| "eval_steps_per_second": 2.01, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 685, | |
| "total_flos": 1.0472781231601746e+18, | |
| "train_loss": 2.151051264783762, | |
| "train_runtime": 5341.9856, | |
| "train_samples_per_second": 2.052, | |
| "train_steps_per_second": 0.128 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 685, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 1.0472781231601746e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |