| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 100, | |
| "global_step": 1250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 6.320438861846924, | |
| "learning_rate": 2.99952628392495e-06, | |
| "loss": 1.3054, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 5.475564956665039, | |
| "learning_rate": 2.9981054349090266e-06, | |
| "loss": 0.6821, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.048, | |
| "grad_norm": 5.251683235168457, | |
| "learning_rate": 2.995738350390921e-06, | |
| "loss": 0.6935, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 5.057413578033447, | |
| "learning_rate": 2.9924265254719506e-06, | |
| "loss": 0.6776, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 4.680449962615967, | |
| "learning_rate": 2.988172051971717e-06, | |
| "loss": 0.6582, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 5.521228313446045, | |
| "learning_rate": 2.982977617106871e-06, | |
| "loss": 0.6972, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.112, | |
| "grad_norm": 4.839247226715088, | |
| "learning_rate": 2.9768465017938084e-06, | |
| "loss": 0.6749, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 5.12637996673584, | |
| "learning_rate": 2.9697825785763704e-06, | |
| "loss": 0.6863, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 4.7851152420043945, | |
| "learning_rate": 2.961790309179866e-06, | |
| "loss": 0.6785, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.761219501495361, | |
| "learning_rate": 2.9528747416929465e-06, | |
| "loss": 0.6727, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.6868906021118164, | |
| "eval_runtime": 8.6852, | |
| "eval_samples_per_second": 230.276, | |
| "eval_steps_per_second": 14.392, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.176, | |
| "grad_norm": 4.815207004547119, | |
| "learning_rate": 2.943041507379129e-06, | |
| "loss": 0.6787, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 4.663937568664551, | |
| "learning_rate": 2.9322968171199645e-06, | |
| "loss": 0.6772, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.208, | |
| "grad_norm": 5.208789348602295, | |
| "learning_rate": 2.9206474574921165e-06, | |
| "loss": 0.656, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 4.607890605926514, | |
| "learning_rate": 2.9081007864808113e-06, | |
| "loss": 0.6792, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 5.205310821533203, | |
| "learning_rate": 2.894664728832377e-06, | |
| "loss": 0.6989, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 4.700341701507568, | |
| "learning_rate": 2.8803477710488056e-06, | |
| "loss": 0.6673, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.272, | |
| "grad_norm": 4.271303653717041, | |
| "learning_rate": 2.8651589560274937e-06, | |
| "loss": 0.6743, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 4.5935468673706055, | |
| "learning_rate": 2.8491078773495566e-06, | |
| "loss": 0.6634, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.304, | |
| "grad_norm": 4.505198955535889, | |
| "learning_rate": 2.832204673220317e-06, | |
| "loss": 0.6487, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 4.946630477905273, | |
| "learning_rate": 2.814460020065795e-06, | |
| "loss": 0.6056, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.6831114292144775, | |
| "eval_runtime": 9.3989, | |
| "eval_samples_per_second": 212.791, | |
| "eval_steps_per_second": 13.299, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.336, | |
| "grad_norm": 4.913090229034424, | |
| "learning_rate": 2.795885125789253e-06, | |
| "loss": 0.652, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 4.906637191772461, | |
| "learning_rate": 2.776491722692038e-06, | |
| "loss": 0.6599, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.368, | |
| "grad_norm": 4.214770317077637, | |
| "learning_rate": 2.756292060063213e-06, | |
| "loss": 0.6539, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 4.808592796325684, | |
| "learning_rate": 2.735298896442641e-06, | |
| "loss": 0.6666, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.6029839515686035, | |
| "learning_rate": 2.713525491562421e-06, | |
| "loss": 0.6924, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 4.828456878662109, | |
| "learning_rate": 2.690985597971753e-06, | |
| "loss": 0.6937, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.432, | |
| "grad_norm": 4.152997970581055, | |
| "learning_rate": 2.6676934523505355e-06, | |
| "loss": 0.6664, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 4.614770412445068, | |
| "learning_rate": 2.643663766517172e-06, | |
| "loss": 0.6722, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.464, | |
| "grad_norm": 4.493509292602539, | |
| "learning_rate": 2.6189117181362736e-06, | |
| "loss": 0.6689, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.5999274253845215, | |
| "learning_rate": 2.5934529411321173e-06, | |
| "loss": 0.7033, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.6796725988388062, | |
| "eval_runtime": 9.6974, | |
| "eval_samples_per_second": 206.241, | |
| "eval_steps_per_second": 12.89, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.496, | |
| "grad_norm": 4.2827935218811035, | |
| "learning_rate": 2.5673035158139285e-06, | |
| "loss": 0.6596, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 4.7542805671691895, | |
| "learning_rate": 2.5404799587192076e-06, | |
| "loss": 0.6549, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.528, | |
| "grad_norm": 4.7150492668151855, | |
| "learning_rate": 2.5129992121815365e-06, | |
| "loss": 0.6757, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 4.777836322784424, | |
| "learning_rate": 2.484878633629435e-06, | |
| "loss": 0.6806, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 4.323108196258545, | |
| "learning_rate": 2.456135984623035e-06, | |
| "loss": 0.6629, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.968432426452637, | |
| "learning_rate": 2.4267894196355018e-06, | |
| "loss": 0.6858, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.592, | |
| "grad_norm": 4.666640758514404, | |
| "learning_rate": 2.3968574745862785e-06, | |
| "loss": 0.665, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 4.655545234680176, | |
| "learning_rate": 2.3663590551334015e-06, | |
| "loss": 0.6661, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.624, | |
| "grad_norm": 4.554004669189453, | |
| "learning_rate": 2.3353134247322823e-06, | |
| "loss": 0.6559, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 4.362437725067139, | |
| "learning_rate": 2.303740192468495e-06, | |
| "loss": 0.6786, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.6770616173744202, | |
| "eval_runtime": 7.3123, | |
| "eval_samples_per_second": 273.512, | |
| "eval_steps_per_second": 17.095, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.656, | |
| "grad_norm": 4.237730503082275, | |
| "learning_rate": 2.2716593006722595e-06, | |
| "loss": 0.6456, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.062387943267822, | |
| "learning_rate": 2.2390910123224374e-06, | |
| "loss": 0.6767, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.688, | |
| "grad_norm": 4.175485134124756, | |
| "learning_rate": 2.2060558982479992e-06, | |
| "loss": 0.6389, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 4.427236080169678, | |
| "learning_rate": 2.1725748241350487e-06, | |
| "loss": 0.6403, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 4.3691511154174805, | |
| "learning_rate": 2.138668937347609e-06, | |
| "loss": 0.6559, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 4.3593430519104, | |
| "learning_rate": 2.1043596535704943e-06, | |
| "loss": 0.6787, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.752, | |
| "grad_norm": 4.350930213928223, | |
| "learning_rate": 2.069668643282702e-06, | |
| "loss": 0.6975, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 4.215636730194092, | |
| "learning_rate": 2.034617818069876e-06, | |
| "loss": 0.684, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.784, | |
| "grad_norm": 3.999462127685547, | |
| "learning_rate": 1.99922931678448e-06, | |
| "loss": 0.6607, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.186202526092529, | |
| "learning_rate": 1.963525491562421e-06, | |
| "loss": 0.6476, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.6736403107643127, | |
| "eval_runtime": 7.3017, | |
| "eval_samples_per_second": 273.908, | |
| "eval_steps_per_second": 17.119, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.816, | |
| "grad_norm": 4.060866355895996, | |
| "learning_rate": 1.927528893704964e-06, | |
| "loss": 0.6591, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 4.125776767730713, | |
| "learning_rate": 1.8912622594348455e-06, | |
| "loss": 0.6922, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.848, | |
| "grad_norm": 3.8623437881469727, | |
| "learning_rate": 1.8547484955355872e-06, | |
| "loss": 0.6513, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 4.550040245056152, | |
| "learning_rate": 1.8180106648830824e-06, | |
| "loss": 0.663, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 4.267393589019775, | |
| "learning_rate": 1.7810719718785873e-06, | |
| "loss": 0.6685, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 4.428436756134033, | |
| "learning_rate": 1.7439557477923257e-06, | |
| "loss": 0.7051, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.912, | |
| "grad_norm": 4.695955276489258, | |
| "learning_rate": 1.706685436026957e-06, | |
| "loss": 0.6251, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 4.578221797943115, | |
| "learning_rate": 1.6692845773102223e-06, | |
| "loss": 0.6569, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.944, | |
| "grad_norm": 4.15753173828125, | |
| "learning_rate": 1.6317767948261151e-06, | |
| "loss": 0.6451, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.946852922439575, | |
| "learning_rate": 1.5941857792939703e-06, | |
| "loss": 0.6562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.6707749962806702, | |
| "eval_runtime": 7.4417, | |
| "eval_samples_per_second": 268.756, | |
| "eval_steps_per_second": 16.797, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.976, | |
| "grad_norm": 4.214157581329346, | |
| "learning_rate": 1.556535274004902e-06, | |
| "loss": 0.6556, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 4.3779616355896, | |
| "learning_rate": 1.518849059825029e-06, | |
| "loss": 0.6764, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.008, | |
| "grad_norm": 3.7831478118896484, | |
| "learning_rate": 1.481150940174971e-06, | |
| "loss": 0.5361, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 4.746358394622803, | |
| "learning_rate": 1.4434647259950982e-06, | |
| "loss": 0.4773, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 4.30097770690918, | |
| "learning_rate": 1.40581422070603e-06, | |
| "loss": 0.4829, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 4.781858444213867, | |
| "learning_rate": 1.3682232051738854e-06, | |
| "loss": 0.4635, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.072, | |
| "grad_norm": 4.712700366973877, | |
| "learning_rate": 1.3307154226897775e-06, | |
| "loss": 0.4775, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.088, | |
| "grad_norm": 4.640942573547363, | |
| "learning_rate": 1.293314563973043e-06, | |
| "loss": 0.4421, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.104, | |
| "grad_norm": 5.104272365570068, | |
| "learning_rate": 1.2560442522076746e-06, | |
| "loss": 0.446, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 5.069291591644287, | |
| "learning_rate": 1.2189280281214128e-06, | |
| "loss": 0.461, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.7041329145431519, | |
| "eval_runtime": 7.9231, | |
| "eval_samples_per_second": 252.428, | |
| "eval_steps_per_second": 15.777, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1360000000000001, | |
| "grad_norm": 5.07060432434082, | |
| "learning_rate": 1.1819893351169183e-06, | |
| "loss": 0.4614, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 6.156511306762695, | |
| "learning_rate": 1.1452515044644133e-06, | |
| "loss": 0.4785, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.168, | |
| "grad_norm": 5.145558834075928, | |
| "learning_rate": 1.108737740565155e-06, | |
| "loss": 0.4463, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.184, | |
| "grad_norm": 4.276706218719482, | |
| "learning_rate": 1.0724711062950359e-06, | |
| "loss": 0.4595, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 5.321038722991943, | |
| "learning_rate": 1.036474508437579e-06, | |
| "loss": 0.4687, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.216, | |
| "grad_norm": 4.610326766967773, | |
| "learning_rate": 1.0007706832155202e-06, | |
| "loss": 0.4293, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.232, | |
| "grad_norm": 5.0618462562561035, | |
| "learning_rate": 9.65382181930124e-07, | |
| "loss": 0.4636, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 4.6133036613464355, | |
| "learning_rate": 9.303313567172986e-07, | |
| "loss": 0.4578, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.264, | |
| "grad_norm": 4.833690166473389, | |
| "learning_rate": 8.956403464295061e-07, | |
| "loss": 0.4511, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 4.997283935546875, | |
| "learning_rate": 8.613310626523911e-07, | |
| "loss": 0.4578, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.7093445062637329, | |
| "eval_runtime": 9.5227, | |
| "eval_samples_per_second": 210.025, | |
| "eval_steps_per_second": 13.127, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.296, | |
| "grad_norm": 5.087992191314697, | |
| "learning_rate": 8.274251758649519e-07, | |
| "loss": 0.4858, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.312, | |
| "grad_norm": 5.305168628692627, | |
| "learning_rate": 7.939441017520012e-07, | |
| "loss": 0.4756, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.328, | |
| "grad_norm": 4.691494464874268, | |
| "learning_rate": 7.609089876775628e-07, | |
| "loss": 0.4593, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 6.369288921356201, | |
| "learning_rate": 7.283406993277403e-07, | |
| "loss": 0.4461, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 4.588751792907715, | |
| "learning_rate": 6.962598075315047e-07, | |
| "loss": 0.4569, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.376, | |
| "grad_norm": 4.815459728240967, | |
| "learning_rate": 6.646865752677186e-07, | |
| "loss": 0.4209, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.392, | |
| "grad_norm": 4.491628170013428, | |
| "learning_rate": 6.336409448665989e-07, | |
| "loss": 0.4768, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 5.085056304931641, | |
| "learning_rate": 6.031425254137223e-07, | |
| "loss": 0.4539, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.424, | |
| "grad_norm": 4.508426189422607, | |
| "learning_rate": 5.732105803644987e-07, | |
| "loss": 0.4588, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 4.900942325592041, | |
| "learning_rate": 5.438640153769653e-07, | |
| "loss": 0.4817, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 0.7055138945579529, | |
| "eval_runtime": 9.5816, | |
| "eval_samples_per_second": 208.733, | |
| "eval_steps_per_second": 13.046, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.456, | |
| "grad_norm": 5.169991970062256, | |
| "learning_rate": 5.151213663705655e-07, | |
| "loss": 0.4608, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.472, | |
| "grad_norm": 5.345458030700684, | |
| "learning_rate": 4.870007878184633e-07, | |
| "loss": 0.4687, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.488, | |
| "grad_norm": 5.129513263702393, | |
| "learning_rate": 4.5952004128079276e-07, | |
| "loss": 0.4677, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.504, | |
| "grad_norm": 4.392848491668701, | |
| "learning_rate": 4.3269648418607197e-07, | |
| "loss": 0.4612, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 4.760442733764648, | |
| "learning_rate": 4.06547058867883e-07, | |
| "loss": 0.4864, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 5.239096164703369, | |
| "learning_rate": 3.8108828186372685e-07, | |
| "loss": 0.4576, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.552, | |
| "grad_norm": 4.897254467010498, | |
| "learning_rate": 3.56336233482828e-07, | |
| "loss": 0.4518, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.568, | |
| "grad_norm": 5.0241804122924805, | |
| "learning_rate": 3.32306547649465e-07, | |
| "loss": 0.454, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.584, | |
| "grad_norm": 5.055388450622559, | |
| "learning_rate": 3.0901440202824693e-07, | |
| "loss": 0.4808, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.961503982543945, | |
| "learning_rate": 2.86474508437579e-07, | |
| "loss": 0.4324, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.7080456614494324, | |
| "eval_runtime": 7.9696, | |
| "eval_samples_per_second": 250.955, | |
| "eval_steps_per_second": 15.685, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.616, | |
| "grad_norm": 4.625386714935303, | |
| "learning_rate": 2.647011035573588e-07, | |
| "loss": 0.4707, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 5.237055778503418, | |
| "learning_rate": 2.437079399367875e-07, | |
| "loss": 0.4575, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.6480000000000001, | |
| "grad_norm": 5.412177562713623, | |
| "learning_rate": 2.235082773079624e-07, | |
| "loss": 0.4786, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 5.122972011566162, | |
| "learning_rate": 2.0411487421074708e-07, | |
| "loss": 0.4961, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 4.920769214630127, | |
| "learning_rate": 1.8553997993420495e-07, | |
| "loss": 0.4494, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.696, | |
| "grad_norm": 4.990310192108154, | |
| "learning_rate": 1.6779532677968329e-07, | |
| "loss": 0.4713, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.712, | |
| "grad_norm": 4.39508581161499, | |
| "learning_rate": 1.508921226504434e-07, | |
| "loss": 0.4633, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 5.249621391296387, | |
| "learning_rate": 1.348410439725065e-07, | |
| "loss": 0.4764, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.744, | |
| "grad_norm": 5.508373260498047, | |
| "learning_rate": 1.1965222895119444e-07, | |
| "loss": 0.4793, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 5.088216781616211, | |
| "learning_rate": 1.0533527116762298e-07, | |
| "loss": 0.4693, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.7081010937690735, | |
| "eval_runtime": 7.6853, | |
| "eval_samples_per_second": 260.237, | |
| "eval_steps_per_second": 16.265, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.776, | |
| "grad_norm": 5.710309028625488, | |
| "learning_rate": 9.18992135191889e-08, | |
| "loss": 0.4405, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 4.6689982414245605, | |
| "learning_rate": 7.935254250788366e-08, | |
| "loss": 0.4484, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.808, | |
| "grad_norm": 4.682315349578857, | |
| "learning_rate": 6.770318288003558e-08, | |
| "loss": 0.4458, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 5.43618106842041, | |
| "learning_rate": 5.6958492620871105e-08, | |
| "loss": 0.4914, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 4.98464822769165, | |
| "learning_rate": 4.712525830705339e-08, | |
| "loss": 0.4567, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.8559999999999999, | |
| "grad_norm": 4.657787799835205, | |
| "learning_rate": 3.820969082013415e-08, | |
| "loss": 0.452, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.8719999999999999, | |
| "grad_norm": 4.8955559730529785, | |
| "learning_rate": 3.021742142362971e-08, | |
| "loss": 0.463, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.888, | |
| "grad_norm": 5.363377094268799, | |
| "learning_rate": 2.3153498206192002e-08, | |
| "loss": 0.4489, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.904, | |
| "grad_norm": 4.830941677093506, | |
| "learning_rate": 1.7022382893129074e-08, | |
| "loss": 0.4594, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 5.2024827003479, | |
| "learning_rate": 1.1827948028283353e-08, | |
| "loss": 0.4475, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.707037091255188, | |
| "eval_runtime": 7.654, | |
| "eval_samples_per_second": 261.3, | |
| "eval_steps_per_second": 16.331, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.936, | |
| "grad_norm": 4.656322956085205, | |
| "learning_rate": 7.57347452804974e-09, | |
| "loss": 0.4583, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.952, | |
| "grad_norm": 5.396835803985596, | |
| "learning_rate": 4.261649609079099e-09, | |
| "loss": 0.4136, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.968, | |
| "grad_norm": 4.812420845031738, | |
| "learning_rate": 1.8945650909737986e-09, | |
| "loss": 0.4706, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.984, | |
| "grad_norm": 4.871044635772705, | |
| "learning_rate": 4.737160750500902e-10, | |
| "loss": 0.4637, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.772403240203857, | |
| "learning_rate": 0.0, | |
| "loss": 0.4707, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 1250, | |
| "total_flos": 4.566411071245517e+16, | |
| "train_loss": 0.5696872653961181, | |
| "train_runtime": 1519.873, | |
| "train_samples_per_second": 13.159, | |
| "train_steps_per_second": 0.822 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.566411071245517e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |