| { | |
| "best_metric": 0.1352197825908661, | |
| "best_model_checkpoint": "saves/psy-course-info/MentaLLaMA-chat-7B/train/fold4/checkpoint-120", | |
| "epoch": 4.938271604938271, | |
| "eval_steps": 10, | |
| "global_step": 125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03950617283950617, | |
| "grad_norm": 0.8409174084663391, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.0295, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.870398998260498, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0028, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 0.9369924068450928, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 1.1596, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.9924418330192566, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.0579, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.9619687795639038, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.1167, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 1.0264861583709717, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.9341, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 1.146074652671814, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 1.055, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.1457151174545288, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.9711, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.3396611213684082, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.975, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 1.279062271118164, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.7643, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.6836758255958557, | |
| "eval_runtime": 23.2076, | |
| "eval_samples_per_second": 1.939, | |
| "eval_steps_per_second": 1.939, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.2532916069030762, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.6847, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.8249327540397644, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.506, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.7587673664093018, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5469, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.5774843096733093, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.2867, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.5543655753135681, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.4322, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.5063629746437073, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.3372, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.5105109810829163, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.3505, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.4840794503688812, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.3899, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.4316415786743164, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.2702, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.4387108385562897, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.2557, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.2727590799331665, | |
| "eval_runtime": 23.2249, | |
| "eval_samples_per_second": 1.938, | |
| "eval_steps_per_second": 1.938, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.4757551848888397, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.2198, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.4873587489128113, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.1807, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.4356836974620819, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.2483, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.4757295548915863, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.2682, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.413959801197052, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.2083, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.8613407015800476, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.3389, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.4239979684352875, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.291, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.31506332755088806, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1775, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.2654768228530884, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.152, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.4273090362548828, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1755, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.20793814957141876, | |
| "eval_runtime": 23.189, | |
| "eval_samples_per_second": 1.941, | |
| "eval_steps_per_second": 1.941, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.42402613162994385, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1678, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.34952279925346375, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.1766, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.44134724140167236, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1622, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.36546987295150757, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1976, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.3772576153278351, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1537, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.3560152053833008, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1663, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.3635655641555786, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.2161, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.36557117104530334, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1731, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.39808523654937744, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.1756, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.48609480261802673, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.1595, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.17323733866214752, | |
| "eval_runtime": 23.2031, | |
| "eval_samples_per_second": 1.939, | |
| "eval_steps_per_second": 1.939, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.38207197189331055, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.176, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.4270172119140625, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.2371, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.3736412525177002, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1561, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.31712430715560913, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.1723, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.2984461784362793, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.1075, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.3510105609893799, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.156, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.3386427164077759, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.1634, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.37029823660850525, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1523, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.4036928415298462, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.1635, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.32910314202308655, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1537, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.15786227583885193, | |
| "eval_runtime": 23.2492, | |
| "eval_samples_per_second": 1.936, | |
| "eval_steps_per_second": 1.936, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.7916931509971619, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.3965, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.3477286100387573, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.1523, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.32445675134658813, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.137, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.29507413506507874, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.1242, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.3310132622718811, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1237, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.2610471844673157, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.1222, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.30773240327835083, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.164, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.4374106824398041, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1528, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.3064025640487671, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.144, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.28123733401298523, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.11, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.14884930849075317, | |
| "eval_runtime": 23.205, | |
| "eval_samples_per_second": 1.939, | |
| "eval_steps_per_second": 1.939, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.3549681603908539, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.1111, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.30231916904449463, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.1102, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.3114653527736664, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0997, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.42086660861968994, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.1564, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.3921828269958496, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1425, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.38940221071243286, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.1484, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.3512428402900696, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.1111, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.4615846574306488, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.1685, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.42115938663482666, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0993, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.2838325798511505, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0938, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.14260053634643555, | |
| "eval_runtime": 23.2387, | |
| "eval_samples_per_second": 1.936, | |
| "eval_steps_per_second": 1.936, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.2949417233467102, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.1324, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.3120761513710022, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0938, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.34445720911026, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.1357, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.42597633600234985, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.1564, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.340300053358078, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.1296, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.7590665221214294, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1698, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.3389676511287689, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0836, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.36551737785339355, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.1218, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.34780970215797424, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.1199, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.2959098517894745, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.1043, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.13685083389282227, | |
| "eval_runtime": 23.2283, | |
| "eval_samples_per_second": 1.937, | |
| "eval_steps_per_second": 1.937, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.32270997762680054, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.101, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.3170534074306488, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0932, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.30391669273376465, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0852, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.37706097960472107, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.1503, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.2684533894062042, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.1102, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.3799731433391571, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.107, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.2965794801712036, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0981, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.3132217526435852, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.1118, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.33298760652542114, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0979, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.3249424695968628, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0975, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.13782529532909393, | |
| "eval_runtime": 23.1987, | |
| "eval_samples_per_second": 1.94, | |
| "eval_steps_per_second": 1.94, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.36549893021583557, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.098, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.2974507510662079, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.1245, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.38176771998405457, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.1301, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.4843205213546753, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.119, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.346051961183548, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0937, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.36730319261550903, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0997, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.40472662448883057, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1287, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.29217612743377686, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0965, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.3884892761707306, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0966, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.3804306089878082, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.105, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.1368541270494461, | |
| "eval_runtime": 23.2258, | |
| "eval_samples_per_second": 1.938, | |
| "eval_steps_per_second": 1.938, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.4681399166584015, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.111, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.5917119383811951, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1772, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.2791345417499542, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.1024, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.3107740879058838, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0771, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.3112463355064392, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0983, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.2696148455142975, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0606, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.37494152784347534, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.1265, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.28171834349632263, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.075, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.3216496706008911, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.1109, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.27308768033981323, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0978, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.13527800142765045, | |
| "eval_runtime": 23.2336, | |
| "eval_samples_per_second": 1.937, | |
| "eval_steps_per_second": 1.937, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.3138827383518219, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.1116, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.3307981491088867, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0958, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.3537079095840454, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0998, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.35708051919937134, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0914, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.36246854066848755, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0763, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.3086746037006378, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.1002, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.3442056179046631, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.1255, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.3429384231567383, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0732, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.3529020845890045, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.1143, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.38459232449531555, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.1241, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.1352197825908661, | |
| "eval_runtime": 23.2566, | |
| "eval_samples_per_second": 1.935, | |
| "eval_steps_per_second": 1.935, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.35208141803741455, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.1006, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.38920432329177856, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0992, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.34282100200653076, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.079, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.41786396503448486, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0859, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.3213942348957062, | |
| "learning_rate": 0.0, | |
| "loss": 0.063, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.380647176790999e+17, | |
| "train_loss": 0.228071633040905, | |
| "train_runtime": 4018.0339, | |
| "train_samples_per_second": 0.504, | |
| "train_steps_per_second": 0.031 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.380647176790999e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |