| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.857142857142858, | |
| "eval_steps": 500, | |
| "global_step": 780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01904761904761905, | |
| "grad_norm": 3.15625, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 3.3205, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 3.3126, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 2.609375, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 3.2719, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 2.828125, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 3.1205, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 5.128205128205128e-05, | |
| "loss": 2.8114, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 2.125, | |
| "learning_rate": 6.410256410256412e-05, | |
| "loss": 2.5384, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 6.03125, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 2.3201, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 8.974358974358975e-05, | |
| "loss": 2.1053, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 0.00010256410256410256, | |
| "loss": 1.8717, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 1.6592, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 0.00012820512820512823, | |
| "loss": 1.518, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9904761904761905, | |
| "eval_loss": 2.7708818912506104, | |
| "eval_runtime": 0.4987, | |
| "eval_samples_per_second": 42.112, | |
| "eval_steps_per_second": 2.005, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.00014102564102564104, | |
| "loss": 1.426, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.3513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 0.75390625, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 1.3036, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.703125, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 1.2627, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.2335, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.00019999599453798524, | |
| "loss": 1.2094, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 0.0001999509367752813, | |
| "loss": 1.1957, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 0.00019985583705641418, | |
| "loss": 1.1772, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.0001997107429945041, | |
| "loss": 1.1595, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 0.0001995157272330992, | |
| "loss": 1.1451, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 1.1423, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.659493923187256, | |
| "eval_runtime": 0.4877, | |
| "eval_samples_per_second": 43.055, | |
| "eval_steps_per_second": 2.05, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 0.00019897634610740287, | |
| "loss": 1.1131, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 0.00019863225079247285, | |
| "loss": 1.1121, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.00019823877374156647, | |
| "loss": 1.103, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 0.00019779611195495177, | |
| "loss": 1.1023, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 1.086, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.00019676414519013781, | |
| "loss": 1.0886, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.0001961753568817896, | |
| "loss": 1.0818, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 0.0001955384169187563, | |
| "loss": 1.0799, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 0.00019485364419471454, | |
| "loss": 1.0792, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.00019412138155154002, | |
| "loss": 1.0681, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.9904761904761905, | |
| "eval_loss": 2.6406373977661133, | |
| "eval_runtime": 0.4936, | |
| "eval_samples_per_second": 42.54, | |
| "eval_steps_per_second": 2.026, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.0001933419956076584, | |
| "loss": 1.0607, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 0.00019251587657449236, | |
| "loss": 1.0467, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00019164343806109632, | |
| "loss": 1.0429, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.00019072511686707663, | |
| "loss": 1.0477, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 0.0001897613727639014, | |
| "loss": 1.0407, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.390625, | |
| "learning_rate": 0.00018875268826470872, | |
| "loss": 1.0408, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.00018769956838272936, | |
| "loss": 1.0392, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.0393, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00018546215349560203, | |
| "loss": 1.0385, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00018427897868623534, | |
| "loss": 1.0383, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 0.00018305360832480117, | |
| "loss": 1.0335, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.642651319503784, | |
| "eval_runtime": 0.4862, | |
| "eval_samples_per_second": 43.193, | |
| "eval_steps_per_second": 2.057, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.00018178665591160172, | |
| "loss": 1.0188, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.00018047875576562557, | |
| "loss": 1.0134, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.0001791305627069662, | |
| "loss": 1.0109, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.0001777427517289766, | |
| "loss": 1.009, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00017631601766032336, | |
| "loss": 1.0116, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 1.0067, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 0.0001733486566452446, | |
| "loss": 1.0112, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 0.0001718095153532274, | |
| "loss": 1.0096, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 0.00017023442153554777, | |
| "loss": 1.0046, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 1.0079, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.9904761904761905, | |
| "eval_loss": 2.645949363708496, | |
| "eval_runtime": 0.4992, | |
| "eval_samples_per_second": 42.069, | |
| "eval_steps_per_second": 2.003, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.00016697954830722868, | |
| "loss": 0.9948, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.34375, | |
| "learning_rate": 0.0001653013984983585, | |
| "loss": 0.9868, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 0.0001635905545514795, | |
| "loss": 0.9845, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 0.0001618478730266255, | |
| "loss": 0.9812, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 0.0001600742264237979, | |
| "loss": 0.9887, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.00015827050274613513, | |
| "loss": 0.9863, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.0001564376050553205, | |
| "loss": 0.9862, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.00015457645101945046, | |
| "loss": 0.9865, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00015268797245359035, | |
| "loss": 0.9929, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0001507731148532468, | |
| "loss": 0.9868, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00014883283692099112, | |
| "loss": 0.9837, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.657383441925049, | |
| "eval_runtime": 0.4893, | |
| "eval_samples_per_second": 42.915, | |
| "eval_steps_per_second": 2.044, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.00014686811008647038, | |
| "loss": 0.9669, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.00014487991802004623, | |
| "loss": 0.9674, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.00014286925614030542, | |
| "loss": 0.9712, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.00014083713111568842, | |
| "loss": 0.9663, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.462890625, | |
| "learning_rate": 0.0001387845603604855, | |
| "loss": 0.9674, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.00013671257152545277, | |
| "loss": 0.9732, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 0.00013462220198330328, | |
| "loss": 0.9743, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.761904761904762, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.0001325144983093305, | |
| "loss": 0.9704, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0001303905157574247, | |
| "loss": 0.9617, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.9523809523809526, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 0.0001282513177317437, | |
| "loss": 0.966, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 6.9904761904761905, | |
| "eval_loss": 2.6700327396392822, | |
| "eval_runtime": 0.5365, | |
| "eval_samples_per_second": 39.146, | |
| "eval_steps_per_second": 1.864, | |
| "step": 367 | |
| }, | |
| { | |
| "epoch": 7.0476190476190474, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.00012609797525430373, | |
| "loss": 0.9592, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.0001239315664287558, | |
| "loss": 0.9535, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 7.238095238095238, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00012175317590061674, | |
| "loss": 0.9528, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.333333333333333, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 0.00011956389431422507, | |
| "loss": 0.9572, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.428571428571429, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.9461, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 7.523809523809524, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.00011515704725912926, | |
| "loss": 0.9513, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.00011294168814540553, | |
| "loss": 0.9566, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.714285714285714, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.00011071984957874479, | |
| "loss": 0.9594, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 7.809523809523809, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.00010849264395640649, | |
| "loss": 0.9554, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 7.904761904761905, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0001062611863627482, | |
| "loss": 0.9479, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 0.00010402659401094152, | |
| "loss": 0.9474, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.6798880100250244, | |
| "eval_runtime": 0.4848, | |
| "eval_samples_per_second": 43.314, | |
| "eval_steps_per_second": 2.063, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.095238095238095, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 0.00010178998568362243, | |
| "loss": 0.9432, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 8.19047619047619, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 9.955248117275566e-05, | |
| "loss": 0.9398, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 8.285714285714286, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 9.73152007189939e-05, | |
| "loss": 0.9348, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 8.380952380952381, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 9.507926445081219e-05, | |
| "loss": 0.9416, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.476190476190476, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 9.284579182369867e-05, | |
| "loss": 0.9352, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 9.061590105968208e-05, | |
| "loss": 0.9427, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.666666666666666, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 8.839070858747697e-05, | |
| "loss": 0.9413, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 8.761904761904763, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 8.617132848352671e-05, | |
| "loss": 0.9409, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 8.395887191422397e-05, | |
| "loss": 0.942, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 8.952380952380953, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 8.175444657958876e-05, | |
| "loss": 0.9406, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 8.99047619047619, | |
| "eval_loss": 2.688331365585327, | |
| "eval_runtime": 0.5482, | |
| "eval_samples_per_second": 38.309, | |
| "eval_steps_per_second": 1.824, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 9.047619047619047, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 7.955915615868111e-05, | |
| "loss": 0.9375, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 7.73740997570278e-05, | |
| "loss": 0.9322, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.238095238095237, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 7.520037135633816e-05, | |
| "loss": 0.9292, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 9.333333333333334, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 7.303905926678564e-05, | |
| "loss": 0.9269, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 9.428571428571429, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 7.089124558212871e-05, | |
| "loss": 0.9331, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 6.875800563794425e-05, | |
| "loss": 0.9344, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.619047619047619, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 6.664040747324437e-05, | |
| "loss": 0.9294, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 6.453951129574644e-05, | |
| "loss": 0.9347, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 9.80952380952381, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 6.245636895106402e-05, | |
| "loss": 0.9282, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 9.904761904761905, | |
| "grad_norm": 0.353515625, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 0.9299, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 5.834750817679606e-05, | |
| "loss": 0.9245, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.6975326538085938, | |
| "eval_runtime": 0.485, | |
| "eval_samples_per_second": 43.303, | |
| "eval_steps_per_second": 2.062, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 10.095238095238095, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 5.6323846910828735e-05, | |
| "loss": 0.9233, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 10.19047619047619, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 5.432205277496327e-05, | |
| "loss": 0.9235, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 10.285714285714286, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 5.234312799786921e-05, | |
| "loss": 0.9194, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 10.380952380952381, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 5.0388063358324134e-05, | |
| "loss": 0.9235, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 10.476190476190476, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 4.845783768916482e-05, | |
| "loss": 0.9217, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.571428571428571, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 4.6553417387219886e-05, | |
| "loss": 0.9243, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 4.467575592946864e-05, | |
| "loss": 0.9262, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 10.761904761904763, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 4.282579339566802e-05, | |
| "loss": 0.9206, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 4.100445599768774e-05, | |
| "loss": 0.9256, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 10.952380952380953, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 3.9212655615787804e-05, | |
| "loss": 0.9208, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 10.99047619047619, | |
| "eval_loss": 2.707897663116455, | |
| "eval_runtime": 0.647, | |
| "eval_samples_per_second": 32.456, | |
| "eval_steps_per_second": 1.546, | |
| "step": 577 | |
| }, | |
| { | |
| "epoch": 11.047619047619047, | |
| "grad_norm": 0.25, | |
| "learning_rate": 3.745128934207225e-05, | |
| "loss": 0.9226, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 11.142857142857142, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.9121, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 11.238095238095237, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 3.402337085960119e-05, | |
| "loss": 0.9155, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 11.333333333333334, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 3.235853489035241e-05, | |
| "loss": 0.9194, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 3.072756464904006e-05, | |
| "loss": 0.9202, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.523809523809524, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 2.9131276705713006e-05, | |
| "loss": 0.9186, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.619047619047619, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 2.7570470266200176e-05, | |
| "loss": 0.9195, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 11.714285714285714, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 2.6045926771976303e-05, | |
| "loss": 0.9223, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 11.80952380952381, | |
| "grad_norm": 0.25, | |
| "learning_rate": 2.4558409508920986e-05, | |
| "loss": 0.9139, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.904761904761905, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 2.3108663225168435e-05, | |
| "loss": 0.9117, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 2.1697413758237784e-05, | |
| "loss": 0.9195, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 2.714789867401123, | |
| "eval_runtime": 0.4906, | |
| "eval_samples_per_second": 42.808, | |
| "eval_steps_per_second": 2.038, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 12.095238095238095, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 2.032536767163141e-05, | |
| "loss": 0.9167, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 12.19047619047619, | |
| "grad_norm": 0.25, | |
| "learning_rate": 1.8993211901083353e-05, | |
| "loss": 0.9144, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 12.285714285714286, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 1.7701613410634365e-05, | |
| "loss": 0.915, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 12.380952380952381, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 0.9153, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 12.476190476190476, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 1.5242654274342894e-05, | |
| "loss": 0.9133, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "grad_norm": 0.25, | |
| "learning_rate": 1.4076524743778319e-05, | |
| "loss": 0.9134, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.666666666666666, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 1.295341410749208e-05, | |
| "loss": 0.9126, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 12.761904761904763, | |
| "grad_norm": 0.25, | |
| "learning_rate": 1.1873884667900125e-05, | |
| "loss": 0.9094, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 1.083847690782972e-05, | |
| "loss": 0.9183, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 12.952380952380953, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 9.8477092199184e-06, | |
| "loss": 0.9212, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 12.99047619047619, | |
| "eval_loss": 2.7153878211975098, | |
| "eval_runtime": 0.5024, | |
| "eval_samples_per_second": 41.796, | |
| "eval_steps_per_second": 1.99, | |
| "step": 682 | |
| }, | |
| { | |
| "epoch": 13.047619047619047, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 8.902077647072881e-06, | |
| "loss": 0.9139, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 13.142857142857142, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 8.002055634117578e-06, | |
| "loss": 0.9123, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 13.238095238095237, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 7.148093790757371e-06, | |
| "loss": 0.9152, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 6.3406196659728465e-06, | |
| "loss": 0.9087, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.428571428571429, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 5.580037533961546e-06, | |
| "loss": 0.9212, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 13.523809523809524, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 4.866728191731829e-06, | |
| "loss": 0.909, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 13.619047619047619, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 4.20104876845111e-06, | |
| "loss": 0.9146, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 3.5833325466437694e-06, | |
| "loss": 0.9107, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 13.80952380952381, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 3.013888795328057e-06, | |
| "loss": 0.9136, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 13.904761904761905, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 2.4930026151759766e-06, | |
| "loss": 0.9147, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 2.0209347957732328e-06, | |
| "loss": 0.9136, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 2.7180798053741455, | |
| "eval_runtime": 0.4838, | |
| "eval_samples_per_second": 43.41, | |
| "eval_steps_per_second": 2.067, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 14.095238095238095, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 1.5979216850509848e-06, | |
| "loss": 0.9092, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 14.19047619047619, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 1.2241750709546917e-06, | |
| "loss": 0.9159, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 8.998820754091531e-07, | |
| "loss": 0.9153, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 14.380952380952381, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 6.25205060633205e-07, | |
| "loss": 0.9114, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 14.476190476190476, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 4.0028154785050063e-07, | |
| "loss": 0.9163, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 14.571428571428571, | |
| "grad_norm": 0.236328125, | |
| "learning_rate": 2.2522414843748618e-07, | |
| "loss": 0.9153, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 14.666666666666666, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 1.0012050754277802e-07, | |
| "loss": 0.912, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.761904761904763, | |
| "grad_norm": 0.25, | |
| "learning_rate": 2.5033260206275277e-08, | |
| "loss": 0.9155, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.0, | |
| "loss": 0.9103, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_loss": 2.71852707862854, | |
| "eval_runtime": 0.4854, | |
| "eval_samples_per_second": 43.26, | |
| "eval_steps_per_second": 2.06, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "step": 780, | |
| "total_flos": 1.2277516310308454e+18, | |
| "train_loss": 1.077159938445458, | |
| "train_runtime": 4175.1629, | |
| "train_samples_per_second": 47.883, | |
| "train_steps_per_second": 0.187 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 780, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 1.2277516310308454e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |