|
{ |
|
"best_metric": 0.36343997716903687, |
|
"best_model_checkpoint": "finetuned-electrical-images/checkpoint-1300", |
|
"epoch": 10.0, |
|
"eval_steps": 100, |
|
"global_step": 2150, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.046511627906976744, |
|
"grad_norm": 2.1625218391418457, |
|
"learning_rate": 0.00019906976744186048, |
|
"loss": 1.5809, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.09302325581395349, |
|
"grad_norm": 2.723552703857422, |
|
"learning_rate": 0.00019813953488372096, |
|
"loss": 1.2765, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.13953488372093023, |
|
"grad_norm": 2.43442702293396, |
|
"learning_rate": 0.0001972093023255814, |
|
"loss": 1.1166, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.18604651162790697, |
|
"grad_norm": 3.831533670425415, |
|
"learning_rate": 0.00019627906976744185, |
|
"loss": 1.0432, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.23255813953488372, |
|
"grad_norm": 1.6888493299484253, |
|
"learning_rate": 0.00019534883720930232, |
|
"loss": 0.8645, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.27906976744186046, |
|
"grad_norm": 3.6029133796691895, |
|
"learning_rate": 0.0001944186046511628, |
|
"loss": 0.8017, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.32558139534883723, |
|
"grad_norm": 3.071220636367798, |
|
"learning_rate": 0.00019348837209302326, |
|
"loss": 0.8024, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.37209302325581395, |
|
"grad_norm": 2.4523532390594482, |
|
"learning_rate": 0.00019255813953488374, |
|
"loss": 0.7961, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.4186046511627907, |
|
"grad_norm": 2.8159303665161133, |
|
"learning_rate": 0.0001916279069767442, |
|
"loss": 0.7968, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.46511627906976744, |
|
"grad_norm": 2.5400731563568115, |
|
"learning_rate": 0.00019069767441860466, |
|
"loss": 0.7236, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.46511627906976744, |
|
"eval_accuracy": 0.8102310231023102, |
|
"eval_loss": 0.6395960450172424, |
|
"eval_runtime": 8.3162, |
|
"eval_samples_per_second": 72.87, |
|
"eval_steps_per_second": 9.139, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.5116279069767442, |
|
"grad_norm": 5.171610355377197, |
|
"learning_rate": 0.00018976744186046513, |
|
"loss": 0.6643, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.5581395348837209, |
|
"grad_norm": 5.763847827911377, |
|
"learning_rate": 0.00018883720930232557, |
|
"loss": 0.6579, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.6046511627906976, |
|
"grad_norm": 2.429523229598999, |
|
"learning_rate": 0.00018790697674418605, |
|
"loss": 0.5872, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.6511627906976745, |
|
"grad_norm": 2.6132380962371826, |
|
"learning_rate": 0.00018697674418604652, |
|
"loss": 0.6796, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.6976744186046512, |
|
"grad_norm": 3.3510539531707764, |
|
"learning_rate": 0.000186046511627907, |
|
"loss": 0.5738, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.7441860465116279, |
|
"grad_norm": 3.596841812133789, |
|
"learning_rate": 0.00018511627906976744, |
|
"loss": 0.5647, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.7906976744186046, |
|
"grad_norm": 2.6907944679260254, |
|
"learning_rate": 0.0001841860465116279, |
|
"loss": 0.5161, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.8372093023255814, |
|
"grad_norm": 2.8201518058776855, |
|
"learning_rate": 0.00018325581395348838, |
|
"loss": 0.5416, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.8837209302325582, |
|
"grad_norm": 3.49650239944458, |
|
"learning_rate": 0.00018232558139534886, |
|
"loss": 0.5873, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.9302325581395349, |
|
"grad_norm": 1.33900785446167, |
|
"learning_rate": 0.0001813953488372093, |
|
"loss": 0.7243, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.9302325581395349, |
|
"eval_accuracy": 0.8333333333333334, |
|
"eval_loss": 0.5123782753944397, |
|
"eval_runtime": 9.3797, |
|
"eval_samples_per_second": 64.608, |
|
"eval_steps_per_second": 8.103, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.9767441860465116, |
|
"grad_norm": 2.2079968452453613, |
|
"learning_rate": 0.00018046511627906977, |
|
"loss": 0.5871, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 1.0232558139534884, |
|
"grad_norm": 2.7433505058288574, |
|
"learning_rate": 0.00017953488372093025, |
|
"loss": 0.5605, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 1.069767441860465, |
|
"grad_norm": 4.832902908325195, |
|
"learning_rate": 0.0001786046511627907, |
|
"loss": 0.4689, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 1.1162790697674418, |
|
"grad_norm": 3.20340633392334, |
|
"learning_rate": 0.00017767441860465117, |
|
"loss": 0.4492, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 1.1627906976744187, |
|
"grad_norm": 3.986785650253296, |
|
"learning_rate": 0.00017674418604651164, |
|
"loss": 0.477, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.2093023255813953, |
|
"grad_norm": 3.679204225540161, |
|
"learning_rate": 0.0001758139534883721, |
|
"loss": 0.4499, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 1.255813953488372, |
|
"grad_norm": 5.091402530670166, |
|
"learning_rate": 0.00017488372093023258, |
|
"loss": 0.4271, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 1.302325581395349, |
|
"grad_norm": 1.7408082485198975, |
|
"learning_rate": 0.00017395348837209303, |
|
"loss": 0.4805, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.3488372093023255, |
|
"grad_norm": 1.776224136352539, |
|
"learning_rate": 0.00017302325581395348, |
|
"loss": 0.5031, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 1.3953488372093024, |
|
"grad_norm": 1.4141583442687988, |
|
"learning_rate": 0.00017209302325581395, |
|
"loss": 0.4288, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.3953488372093024, |
|
"eval_accuracy": 0.863036303630363, |
|
"eval_loss": 0.45135730504989624, |
|
"eval_runtime": 8.2539, |
|
"eval_samples_per_second": 73.42, |
|
"eval_steps_per_second": 9.208, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.441860465116279, |
|
"grad_norm": 3.1988017559051514, |
|
"learning_rate": 0.00017116279069767442, |
|
"loss": 0.4641, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 1.4883720930232558, |
|
"grad_norm": 3.4049489498138428, |
|
"learning_rate": 0.0001702325581395349, |
|
"loss": 0.5539, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.5348837209302326, |
|
"grad_norm": 2.879892349243164, |
|
"learning_rate": 0.00016930232558139537, |
|
"loss": 0.4818, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 1.5813953488372094, |
|
"grad_norm": 3.1155309677124023, |
|
"learning_rate": 0.00016837209302325584, |
|
"loss": 0.5773, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 1.627906976744186, |
|
"grad_norm": 5.9107866287231445, |
|
"learning_rate": 0.00016744186046511629, |
|
"loss": 0.5665, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 1.6744186046511627, |
|
"grad_norm": 3.242166519165039, |
|
"learning_rate": 0.00016651162790697673, |
|
"loss": 0.4483, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 1.7209302325581395, |
|
"grad_norm": 2.6073851585388184, |
|
"learning_rate": 0.0001655813953488372, |
|
"loss": 0.3956, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 1.7674418604651163, |
|
"grad_norm": 2.7050387859344482, |
|
"learning_rate": 0.00016465116279069768, |
|
"loss": 0.5734, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 1.8139534883720931, |
|
"grad_norm": 4.9598259925842285, |
|
"learning_rate": 0.00016372093023255815, |
|
"loss": 0.6565, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 1.8604651162790697, |
|
"grad_norm": 2.172515630722046, |
|
"learning_rate": 0.00016279069767441862, |
|
"loss": 0.5744, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.8604651162790697, |
|
"eval_accuracy": 0.8102310231023102, |
|
"eval_loss": 0.6154166460037231, |
|
"eval_runtime": 8.1716, |
|
"eval_samples_per_second": 74.159, |
|
"eval_steps_per_second": 9.3, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.9069767441860463, |
|
"grad_norm": 1.5915610790252686, |
|
"learning_rate": 0.00016186046511627907, |
|
"loss": 0.6044, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 1.9534883720930232, |
|
"grad_norm": 3.225754976272583, |
|
"learning_rate": 0.00016093023255813954, |
|
"loss": 0.5832, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 5.616280555725098, |
|
"learning_rate": 0.00016, |
|
"loss": 0.3845, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 2.046511627906977, |
|
"grad_norm": 2.0786819458007812, |
|
"learning_rate": 0.00015906976744186046, |
|
"loss": 0.4349, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 2.0930232558139537, |
|
"grad_norm": 2.8910179138183594, |
|
"learning_rate": 0.00015813953488372093, |
|
"loss": 0.4858, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 2.13953488372093, |
|
"grad_norm": 2.445451259613037, |
|
"learning_rate": 0.0001572093023255814, |
|
"loss": 0.3341, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 2.186046511627907, |
|
"grad_norm": 3.979907751083374, |
|
"learning_rate": 0.00015627906976744188, |
|
"loss": 0.4102, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 2.2325581395348837, |
|
"grad_norm": 3.332590103149414, |
|
"learning_rate": 0.00015534883720930232, |
|
"loss": 0.4107, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 2.2790697674418605, |
|
"grad_norm": 3.54638671875, |
|
"learning_rate": 0.0001544186046511628, |
|
"loss": 0.3151, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 2.3255813953488373, |
|
"grad_norm": 5.199957847595215, |
|
"learning_rate": 0.00015348837209302327, |
|
"loss": 0.4077, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.3255813953488373, |
|
"eval_accuracy": 0.8613861386138614, |
|
"eval_loss": 0.46117761731147766, |
|
"eval_runtime": 8.1068, |
|
"eval_samples_per_second": 74.752, |
|
"eval_steps_per_second": 9.375, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.3720930232558137, |
|
"grad_norm": 4.220390796661377, |
|
"learning_rate": 0.00015255813953488374, |
|
"loss": 0.4136, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 2.4186046511627906, |
|
"grad_norm": 3.7662441730499268, |
|
"learning_rate": 0.0001516279069767442, |
|
"loss": 0.4278, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 2.4651162790697674, |
|
"grad_norm": 3.1768009662628174, |
|
"learning_rate": 0.00015069767441860466, |
|
"loss": 0.3468, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 2.511627906976744, |
|
"grad_norm": 4.637648105621338, |
|
"learning_rate": 0.0001497674418604651, |
|
"loss": 0.4745, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 2.558139534883721, |
|
"grad_norm": 4.119113445281982, |
|
"learning_rate": 0.00014883720930232558, |
|
"loss": 0.3556, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 2.604651162790698, |
|
"grad_norm": 1.932075023651123, |
|
"learning_rate": 0.00014790697674418605, |
|
"loss": 0.3924, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 2.6511627906976747, |
|
"grad_norm": 1.4679527282714844, |
|
"learning_rate": 0.00014697674418604652, |
|
"loss": 0.492, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 2.697674418604651, |
|
"grad_norm": 3.493939161300659, |
|
"learning_rate": 0.000146046511627907, |
|
"loss": 0.4785, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 2.744186046511628, |
|
"grad_norm": 5.127101898193359, |
|
"learning_rate": 0.00014511627906976747, |
|
"loss": 0.3809, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 2.7906976744186047, |
|
"grad_norm": 4.6078925132751465, |
|
"learning_rate": 0.00014418604651162791, |
|
"loss": 0.496, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.7906976744186047, |
|
"eval_accuracy": 0.8729372937293729, |
|
"eval_loss": 0.43590766191482544, |
|
"eval_runtime": 7.9386, |
|
"eval_samples_per_second": 76.336, |
|
"eval_steps_per_second": 9.573, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.8372093023255816, |
|
"grad_norm": 2.777296781539917, |
|
"learning_rate": 0.00014325581395348836, |
|
"loss": 0.3873, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 2.883720930232558, |
|
"grad_norm": 3.4130568504333496, |
|
"learning_rate": 0.00014232558139534883, |
|
"loss": 0.4155, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 2.9302325581395348, |
|
"grad_norm": 3.3363895416259766, |
|
"learning_rate": 0.0001413953488372093, |
|
"loss": 0.4497, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 2.9767441860465116, |
|
"grad_norm": 2.6814212799072266, |
|
"learning_rate": 0.00014046511627906978, |
|
"loss": 0.3669, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 3.0232558139534884, |
|
"grad_norm": 4.29351282119751, |
|
"learning_rate": 0.00013953488372093025, |
|
"loss": 0.3421, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 3.0697674418604652, |
|
"grad_norm": 2.0479118824005127, |
|
"learning_rate": 0.00013860465116279072, |
|
"loss": 0.3297, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 3.116279069767442, |
|
"grad_norm": 1.3133976459503174, |
|
"learning_rate": 0.00013767441860465117, |
|
"loss": 0.2098, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 3.1627906976744184, |
|
"grad_norm": 0.7082284092903137, |
|
"learning_rate": 0.00013674418604651162, |
|
"loss": 0.4151, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 3.2093023255813953, |
|
"grad_norm": 3.3736181259155273, |
|
"learning_rate": 0.0001358139534883721, |
|
"loss": 0.2073, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 3.255813953488372, |
|
"grad_norm": 7.461119174957275, |
|
"learning_rate": 0.00013488372093023256, |
|
"loss": 0.3446, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 3.255813953488372, |
|
"eval_accuracy": 0.8696369636963697, |
|
"eval_loss": 0.42761871218681335, |
|
"eval_runtime": 8.0701, |
|
"eval_samples_per_second": 75.092, |
|
"eval_steps_per_second": 9.417, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 3.302325581395349, |
|
"grad_norm": 1.2921802997589111, |
|
"learning_rate": 0.00013395348837209303, |
|
"loss": 0.3886, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 3.3488372093023258, |
|
"grad_norm": 2.9924850463867188, |
|
"learning_rate": 0.0001330232558139535, |
|
"loss": 0.2564, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 3.395348837209302, |
|
"grad_norm": 3.2593166828155518, |
|
"learning_rate": 0.00013209302325581395, |
|
"loss": 0.404, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 3.441860465116279, |
|
"grad_norm": 2.4605319499969482, |
|
"learning_rate": 0.00013116279069767442, |
|
"loss": 0.5092, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 3.488372093023256, |
|
"grad_norm": 2.1411373615264893, |
|
"learning_rate": 0.0001302325581395349, |
|
"loss": 0.3141, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 3.5348837209302326, |
|
"grad_norm": 3.7920267581939697, |
|
"learning_rate": 0.00012930232558139534, |
|
"loss": 0.2392, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 3.5813953488372094, |
|
"grad_norm": 2.0424487590789795, |
|
"learning_rate": 0.00012837209302325582, |
|
"loss": 0.3916, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 3.6279069767441863, |
|
"grad_norm": 3.782304286956787, |
|
"learning_rate": 0.0001274418604651163, |
|
"loss": 0.2629, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 3.6744186046511627, |
|
"grad_norm": 1.529003381729126, |
|
"learning_rate": 0.00012651162790697676, |
|
"loss": 0.3023, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 3.7209302325581395, |
|
"grad_norm": 5.351986885070801, |
|
"learning_rate": 0.0001255813953488372, |
|
"loss": 0.3347, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 3.7209302325581395, |
|
"eval_accuracy": 0.8795379537953796, |
|
"eval_loss": 0.4259110689163208, |
|
"eval_runtime": 8.618, |
|
"eval_samples_per_second": 70.318, |
|
"eval_steps_per_second": 8.819, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 3.7674418604651163, |
|
"grad_norm": 4.942080974578857, |
|
"learning_rate": 0.00012465116279069768, |
|
"loss": 0.2278, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 3.813953488372093, |
|
"grad_norm": 2.3388428688049316, |
|
"learning_rate": 0.00012372093023255815, |
|
"loss": 0.1785, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 3.8604651162790695, |
|
"grad_norm": 2.0570669174194336, |
|
"learning_rate": 0.00012279069767441863, |
|
"loss": 0.3932, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 3.9069767441860463, |
|
"grad_norm": 5.025542736053467, |
|
"learning_rate": 0.00012186046511627907, |
|
"loss": 0.3624, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 3.953488372093023, |
|
"grad_norm": 2.5320193767547607, |
|
"learning_rate": 0.00012093023255813953, |
|
"loss": 0.2873, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 7.442944526672363, |
|
"learning_rate": 0.00012, |
|
"loss": 0.3133, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 4.046511627906977, |
|
"grad_norm": 5.330275535583496, |
|
"learning_rate": 0.00011906976744186048, |
|
"loss": 0.2818, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 4.093023255813954, |
|
"grad_norm": 4.320654392242432, |
|
"learning_rate": 0.00011813953488372094, |
|
"loss": 0.4261, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 4.1395348837209305, |
|
"grad_norm": 3.309112310409546, |
|
"learning_rate": 0.00011720930232558141, |
|
"loss": 0.2859, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 4.186046511627907, |
|
"grad_norm": 2.0252270698547363, |
|
"learning_rate": 0.00011627906976744187, |
|
"loss": 0.3868, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 4.186046511627907, |
|
"eval_accuracy": 0.8547854785478548, |
|
"eval_loss": 0.46416693925857544, |
|
"eval_runtime": 8.3175, |
|
"eval_samples_per_second": 72.859, |
|
"eval_steps_per_second": 9.137, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 4.232558139534884, |
|
"grad_norm": 0.6425607204437256, |
|
"learning_rate": 0.00011534883720930234, |
|
"loss": 0.224, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 4.27906976744186, |
|
"grad_norm": 1.0721712112426758, |
|
"learning_rate": 0.00011441860465116279, |
|
"loss": 0.2792, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 4.325581395348837, |
|
"grad_norm": 3.4592926502227783, |
|
"learning_rate": 0.00011348837209302326, |
|
"loss": 0.1881, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 4.372093023255814, |
|
"grad_norm": 3.121467351913452, |
|
"learning_rate": 0.00011255813953488372, |
|
"loss": 0.2873, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 4.4186046511627906, |
|
"grad_norm": 1.8312969207763672, |
|
"learning_rate": 0.00011162790697674419, |
|
"loss": 0.2052, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 4.465116279069767, |
|
"grad_norm": 2.4395313262939453, |
|
"learning_rate": 0.00011069767441860466, |
|
"loss": 0.2376, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 4.511627906976744, |
|
"grad_norm": 4.029181480407715, |
|
"learning_rate": 0.00010976744186046512, |
|
"loss": 0.2401, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 4.558139534883721, |
|
"grad_norm": 5.82197904586792, |
|
"learning_rate": 0.0001088372093023256, |
|
"loss": 0.3262, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 4.604651162790698, |
|
"grad_norm": 3.295616388320923, |
|
"learning_rate": 0.00010790697674418607, |
|
"loss": 0.233, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 4.651162790697675, |
|
"grad_norm": 4.5378899574279785, |
|
"learning_rate": 0.00010697674418604651, |
|
"loss": 0.36, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.651162790697675, |
|
"eval_accuracy": 0.8696369636963697, |
|
"eval_loss": 0.4241912066936493, |
|
"eval_runtime": 11.6827, |
|
"eval_samples_per_second": 51.872, |
|
"eval_steps_per_second": 6.505, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.6976744186046515, |
|
"grad_norm": 2.949002504348755, |
|
"learning_rate": 0.00010604651162790697, |
|
"loss": 0.1716, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 4.7441860465116275, |
|
"grad_norm": 3.2645139694213867, |
|
"learning_rate": 0.00010511627906976745, |
|
"loss": 0.2992, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 4.790697674418604, |
|
"grad_norm": 0.3114562928676605, |
|
"learning_rate": 0.0001041860465116279, |
|
"loss": 0.268, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 4.837209302325581, |
|
"grad_norm": 1.4417003393173218, |
|
"learning_rate": 0.00010325581395348838, |
|
"loss": 0.3601, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 4.883720930232558, |
|
"grad_norm": 2.7006003856658936, |
|
"learning_rate": 0.00010232558139534885, |
|
"loss": 0.335, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 4.930232558139535, |
|
"grad_norm": 2.294205665588379, |
|
"learning_rate": 0.00010139534883720931, |
|
"loss": 0.2171, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 4.976744186046512, |
|
"grad_norm": 4.027862071990967, |
|
"learning_rate": 0.00010046511627906978, |
|
"loss": 0.3075, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 5.023255813953488, |
|
"grad_norm": 2.2804319858551025, |
|
"learning_rate": 9.953488372093024e-05, |
|
"loss": 0.2058, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 5.069767441860465, |
|
"grad_norm": 3.193336009979248, |
|
"learning_rate": 9.86046511627907e-05, |
|
"loss": 0.1606, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 5.116279069767442, |
|
"grad_norm": 1.5344799757003784, |
|
"learning_rate": 9.767441860465116e-05, |
|
"loss": 0.295, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 5.116279069767442, |
|
"eval_accuracy": 0.8811881188118812, |
|
"eval_loss": 0.4204191267490387, |
|
"eval_runtime": 7.9394, |
|
"eval_samples_per_second": 76.329, |
|
"eval_steps_per_second": 9.573, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 5.162790697674419, |
|
"grad_norm": 2.80682110786438, |
|
"learning_rate": 9.674418604651163e-05, |
|
"loss": 0.2268, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 5.209302325581396, |
|
"grad_norm": 2.198103904724121, |
|
"learning_rate": 9.58139534883721e-05, |
|
"loss": 0.2226, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 5.2558139534883725, |
|
"grad_norm": 3.6225826740264893, |
|
"learning_rate": 9.488372093023256e-05, |
|
"loss": 0.1485, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 5.3023255813953485, |
|
"grad_norm": 1.2400827407836914, |
|
"learning_rate": 9.395348837209302e-05, |
|
"loss": 0.1442, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 5.348837209302325, |
|
"grad_norm": 1.0701277256011963, |
|
"learning_rate": 9.30232558139535e-05, |
|
"loss": 0.2154, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 5.395348837209302, |
|
"grad_norm": 1.8416105508804321, |
|
"learning_rate": 9.209302325581396e-05, |
|
"loss": 0.2288, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 5.441860465116279, |
|
"grad_norm": 5.05810546875, |
|
"learning_rate": 9.116279069767443e-05, |
|
"loss": 0.3002, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 5.488372093023256, |
|
"grad_norm": 2.076577663421631, |
|
"learning_rate": 9.023255813953489e-05, |
|
"loss": 0.1431, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 5.534883720930233, |
|
"grad_norm": 1.4769999980926514, |
|
"learning_rate": 8.930232558139535e-05, |
|
"loss": 0.2355, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 5.5813953488372094, |
|
"grad_norm": 4.101127624511719, |
|
"learning_rate": 8.837209302325582e-05, |
|
"loss": 0.2342, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 5.5813953488372094, |
|
"eval_accuracy": 0.8910891089108911, |
|
"eval_loss": 0.3933403491973877, |
|
"eval_runtime": 8.1318, |
|
"eval_samples_per_second": 74.523, |
|
"eval_steps_per_second": 9.346, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 5.627906976744186, |
|
"grad_norm": 3.6835029125213623, |
|
"learning_rate": 8.744186046511629e-05, |
|
"loss": 0.221, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 5.674418604651163, |
|
"grad_norm": 5.327223300933838, |
|
"learning_rate": 8.651162790697674e-05, |
|
"loss": 0.2887, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 5.720930232558139, |
|
"grad_norm": 1.522955060005188, |
|
"learning_rate": 8.558139534883721e-05, |
|
"loss": 0.2057, |
|
"step": 1230 |
|
}, |
|
{ |
|
"epoch": 5.767441860465116, |
|
"grad_norm": 5.1004133224487305, |
|
"learning_rate": 8.465116279069768e-05, |
|
"loss": 0.2433, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 5.813953488372093, |
|
"grad_norm": 1.7744640111923218, |
|
"learning_rate": 8.372093023255814e-05, |
|
"loss": 0.1493, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 5.8604651162790695, |
|
"grad_norm": 4.374938488006592, |
|
"learning_rate": 8.27906976744186e-05, |
|
"loss": 0.3513, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 5.906976744186046, |
|
"grad_norm": 2.1579113006591797, |
|
"learning_rate": 8.186046511627907e-05, |
|
"loss": 0.1253, |
|
"step": 1270 |
|
}, |
|
{ |
|
"epoch": 5.953488372093023, |
|
"grad_norm": 3.5076637268066406, |
|
"learning_rate": 8.093023255813953e-05, |
|
"loss": 0.2042, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"grad_norm": 2.294401168823242, |
|
"learning_rate": 8e-05, |
|
"loss": 0.3253, |
|
"step": 1290 |
|
}, |
|
{ |
|
"epoch": 6.046511627906977, |
|
"grad_norm": 0.5659641027450562, |
|
"learning_rate": 7.906976744186047e-05, |
|
"loss": 0.1629, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 6.046511627906977, |
|
"eval_accuracy": 0.8976897689768977, |
|
"eval_loss": 0.36343997716903687, |
|
"eval_runtime": 8.0696, |
|
"eval_samples_per_second": 75.096, |
|
"eval_steps_per_second": 9.418, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 6.093023255813954, |
|
"grad_norm": 3.126523971557617, |
|
"learning_rate": 7.813953488372094e-05, |
|
"loss": 0.1177, |
|
"step": 1310 |
|
}, |
|
{ |
|
"epoch": 6.1395348837209305, |
|
"grad_norm": 7.47089958190918, |
|
"learning_rate": 7.72093023255814e-05, |
|
"loss": 0.2036, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 6.186046511627907, |
|
"grad_norm": 1.5586128234863281, |
|
"learning_rate": 7.627906976744187e-05, |
|
"loss": 0.1308, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 6.232558139534884, |
|
"grad_norm": 2.230147361755371, |
|
"learning_rate": 7.534883720930233e-05, |
|
"loss": 0.2078, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 6.27906976744186, |
|
"grad_norm": 4.571086883544922, |
|
"learning_rate": 7.441860465116279e-05, |
|
"loss": 0.131, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 6.325581395348837, |
|
"grad_norm": 0.7467085719108582, |
|
"learning_rate": 7.348837209302326e-05, |
|
"loss": 0.1465, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 6.372093023255814, |
|
"grad_norm": 1.2842942476272583, |
|
"learning_rate": 7.255813953488373e-05, |
|
"loss": 0.1553, |
|
"step": 1370 |
|
}, |
|
{ |
|
"epoch": 6.4186046511627906, |
|
"grad_norm": 0.23809537291526794, |
|
"learning_rate": 7.162790697674418e-05, |
|
"loss": 0.1314, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 6.465116279069767, |
|
"grad_norm": 8.466991424560547, |
|
"learning_rate": 7.069767441860465e-05, |
|
"loss": 0.2773, |
|
"step": 1390 |
|
}, |
|
{ |
|
"epoch": 6.511627906976744, |
|
"grad_norm": 11.282208442687988, |
|
"learning_rate": 6.976744186046513e-05, |
|
"loss": 0.2041, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 6.511627906976744, |
|
"eval_accuracy": 0.8910891089108911, |
|
"eval_loss": 0.4006832540035248, |
|
"eval_runtime": 10.4521, |
|
"eval_samples_per_second": 57.979, |
|
"eval_steps_per_second": 7.271, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 6.558139534883721, |
|
"grad_norm": 0.22673094272613525, |
|
"learning_rate": 6.883720930232558e-05, |
|
"loss": 0.1841, |
|
"step": 1410 |
|
}, |
|
{ |
|
"epoch": 6.604651162790698, |
|
"grad_norm": 1.8003208637237549, |
|
"learning_rate": 6.790697674418604e-05, |
|
"loss": 0.2147, |
|
"step": 1420 |
|
}, |
|
{ |
|
"epoch": 6.651162790697675, |
|
"grad_norm": 3.2758359909057617, |
|
"learning_rate": 6.697674418604652e-05, |
|
"loss": 0.1884, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 6.6976744186046515, |
|
"grad_norm": 2.0551984310150146, |
|
"learning_rate": 6.604651162790698e-05, |
|
"loss": 0.2099, |
|
"step": 1440 |
|
}, |
|
{ |
|
"epoch": 6.7441860465116275, |
|
"grad_norm": 0.6447792053222656, |
|
"learning_rate": 6.511627906976745e-05, |
|
"loss": 0.114, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 6.790697674418604, |
|
"grad_norm": 0.8234413862228394, |
|
"learning_rate": 6.418604651162791e-05, |
|
"loss": 0.141, |
|
"step": 1460 |
|
}, |
|
{ |
|
"epoch": 6.837209302325581, |
|
"grad_norm": 4.1033406257629395, |
|
"learning_rate": 6.325581395348838e-05, |
|
"loss": 0.1895, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 6.883720930232558, |
|
"grad_norm": 3.889683246612549, |
|
"learning_rate": 6.232558139534884e-05, |
|
"loss": 0.1317, |
|
"step": 1480 |
|
}, |
|
{ |
|
"epoch": 6.930232558139535, |
|
"grad_norm": 4.772053241729736, |
|
"learning_rate": 6.139534883720931e-05, |
|
"loss": 0.2148, |
|
"step": 1490 |
|
}, |
|
{ |
|
"epoch": 6.976744186046512, |
|
"grad_norm": 3.037203788757324, |
|
"learning_rate": 6.0465116279069765e-05, |
|
"loss": 0.1668, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 6.976744186046512, |
|
"eval_accuracy": 0.8927392739273927, |
|
"eval_loss": 0.3843345642089844, |
|
"eval_runtime": 8.3737, |
|
"eval_samples_per_second": 72.37, |
|
"eval_steps_per_second": 9.076, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 7.023255813953488, |
|
"grad_norm": 4.356062889099121, |
|
"learning_rate": 5.953488372093024e-05, |
|
"loss": 0.2231, |
|
"step": 1510 |
|
}, |
|
{ |
|
"epoch": 7.069767441860465, |
|
"grad_norm": 3.485504388809204, |
|
"learning_rate": 5.8604651162790704e-05, |
|
"loss": 0.1391, |
|
"step": 1520 |
|
}, |
|
{ |
|
"epoch": 7.116279069767442, |
|
"grad_norm": 1.0620895624160767, |
|
"learning_rate": 5.767441860465117e-05, |
|
"loss": 0.1369, |
|
"step": 1530 |
|
}, |
|
{ |
|
"epoch": 7.162790697674419, |
|
"grad_norm": 0.46343836188316345, |
|
"learning_rate": 5.674418604651163e-05, |
|
"loss": 0.1639, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 7.209302325581396, |
|
"grad_norm": 5.317925453186035, |
|
"learning_rate": 5.5813953488372095e-05, |
|
"loss": 0.1061, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 7.2558139534883725, |
|
"grad_norm": 2.8796050548553467, |
|
"learning_rate": 5.488372093023256e-05, |
|
"loss": 0.164, |
|
"step": 1560 |
|
}, |
|
{ |
|
"epoch": 7.3023255813953485, |
|
"grad_norm": 2.1574888229370117, |
|
"learning_rate": 5.3953488372093034e-05, |
|
"loss": 0.0833, |
|
"step": 1570 |
|
}, |
|
{ |
|
"epoch": 7.348837209302325, |
|
"grad_norm": 4.602242469787598, |
|
"learning_rate": 5.3023255813953486e-05, |
|
"loss": 0.1002, |
|
"step": 1580 |
|
}, |
|
{ |
|
"epoch": 7.395348837209302, |
|
"grad_norm": 1.2774103879928589, |
|
"learning_rate": 5.209302325581395e-05, |
|
"loss": 0.0654, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 7.441860465116279, |
|
"grad_norm": 0.37218332290649414, |
|
"learning_rate": 5.1162790697674425e-05, |
|
"loss": 0.0976, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 7.441860465116279, |
|
"eval_accuracy": 0.8927392739273927, |
|
"eval_loss": 0.4061937630176544, |
|
"eval_runtime": 8.5468, |
|
"eval_samples_per_second": 70.904, |
|
"eval_steps_per_second": 8.892, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 7.488372093023256, |
|
"grad_norm": 1.125712275505066, |
|
"learning_rate": 5.023255813953489e-05, |
|
"loss": 0.1522, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 7.534883720930233, |
|
"grad_norm": 3.255833625793457, |
|
"learning_rate": 4.930232558139535e-05, |
|
"loss": 0.154, |
|
"step": 1620 |
|
}, |
|
{ |
|
"epoch": 7.5813953488372094, |
|
"grad_norm": 3.5783629417419434, |
|
"learning_rate": 4.8372093023255816e-05, |
|
"loss": 0.1385, |
|
"step": 1630 |
|
}, |
|
{ |
|
"epoch": 7.627906976744186, |
|
"grad_norm": 4.361517429351807, |
|
"learning_rate": 4.744186046511628e-05, |
|
"loss": 0.1105, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 7.674418604651163, |
|
"grad_norm": 1.0743229389190674, |
|
"learning_rate": 4.651162790697675e-05, |
|
"loss": 0.1461, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 7.720930232558139, |
|
"grad_norm": 1.155990481376648, |
|
"learning_rate": 4.5581395348837214e-05, |
|
"loss": 0.1247, |
|
"step": 1660 |
|
}, |
|
{ |
|
"epoch": 7.767441860465116, |
|
"grad_norm": 0.03133929520845413, |
|
"learning_rate": 4.465116279069767e-05, |
|
"loss": 0.0625, |
|
"step": 1670 |
|
}, |
|
{ |
|
"epoch": 7.813953488372093, |
|
"grad_norm": 1.1082956790924072, |
|
"learning_rate": 4.3720930232558146e-05, |
|
"loss": 0.2145, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 7.8604651162790695, |
|
"grad_norm": 2.6899712085723877, |
|
"learning_rate": 4.2790697674418605e-05, |
|
"loss": 0.1182, |
|
"step": 1690 |
|
}, |
|
{ |
|
"epoch": 7.906976744186046, |
|
"grad_norm": 5.237185955047607, |
|
"learning_rate": 4.186046511627907e-05, |
|
"loss": 0.1275, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 7.906976744186046, |
|
"eval_accuracy": 0.8894389438943895, |
|
"eval_loss": 0.3860972821712494, |
|
"eval_runtime": 8.3425, |
|
"eval_samples_per_second": 72.64, |
|
"eval_steps_per_second": 9.11, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 7.953488372093023, |
|
"grad_norm": 1.522539496421814, |
|
"learning_rate": 4.093023255813954e-05, |
|
"loss": 0.1341, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"grad_norm": 0.16910937428474426, |
|
"learning_rate": 4e-05, |
|
"loss": 0.1398, |
|
"step": 1720 |
|
}, |
|
{ |
|
"epoch": 8.046511627906977, |
|
"grad_norm": 3.6039626598358154, |
|
"learning_rate": 3.906976744186047e-05, |
|
"loss": 0.089, |
|
"step": 1730 |
|
}, |
|
{ |
|
"epoch": 8.093023255813954, |
|
"grad_norm": 3.6135542392730713, |
|
"learning_rate": 3.8139534883720935e-05, |
|
"loss": 0.1332, |
|
"step": 1740 |
|
}, |
|
{ |
|
"epoch": 8.13953488372093, |
|
"grad_norm": 2.710392713546753, |
|
"learning_rate": 3.7209302325581394e-05, |
|
"loss": 0.1241, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 8.186046511627907, |
|
"grad_norm": 5.7339067459106445, |
|
"learning_rate": 3.627906976744187e-05, |
|
"loss": 0.1269, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 8.232558139534884, |
|
"grad_norm": 1.990607738494873, |
|
"learning_rate": 3.5348837209302326e-05, |
|
"loss": 0.1111, |
|
"step": 1770 |
|
}, |
|
{ |
|
"epoch": 8.279069767441861, |
|
"grad_norm": 0.09675276279449463, |
|
"learning_rate": 3.441860465116279e-05, |
|
"loss": 0.116, |
|
"step": 1780 |
|
}, |
|
{ |
|
"epoch": 8.325581395348838, |
|
"grad_norm": 0.04001321643590927, |
|
"learning_rate": 3.348837209302326e-05, |
|
"loss": 0.1184, |
|
"step": 1790 |
|
}, |
|
{ |
|
"epoch": 8.372093023255815, |
|
"grad_norm": 0.477737694978714, |
|
"learning_rate": 3.2558139534883724e-05, |
|
"loss": 0.1063, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 8.372093023255815, |
|
"eval_accuracy": 0.8910891089108911, |
|
"eval_loss": 0.40107399225234985, |
|
"eval_runtime": 8.3287, |
|
"eval_samples_per_second": 72.761, |
|
"eval_steps_per_second": 9.125, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 8.418604651162791, |
|
"grad_norm": 0.2791198790073395, |
|
"learning_rate": 3.162790697674419e-05, |
|
"loss": 0.1073, |
|
"step": 1810 |
|
}, |
|
{ |
|
"epoch": 8.465116279069768, |
|
"grad_norm": 0.03783993795514107, |
|
"learning_rate": 3.0697674418604656e-05, |
|
"loss": 0.1584, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 8.511627906976745, |
|
"grad_norm": 0.7814832925796509, |
|
"learning_rate": 2.976744186046512e-05, |
|
"loss": 0.0507, |
|
"step": 1830 |
|
}, |
|
{ |
|
"epoch": 8.55813953488372, |
|
"grad_norm": 0.22078533470630646, |
|
"learning_rate": 2.8837209302325585e-05, |
|
"loss": 0.0539, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 8.604651162790697, |
|
"grad_norm": 2.293942928314209, |
|
"learning_rate": 2.7906976744186048e-05, |
|
"loss": 0.1488, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 8.651162790697674, |
|
"grad_norm": 0.0680939108133316, |
|
"learning_rate": 2.6976744186046517e-05, |
|
"loss": 0.071, |
|
"step": 1860 |
|
}, |
|
{ |
|
"epoch": 8.69767441860465, |
|
"grad_norm": 2.966674327850342, |
|
"learning_rate": 2.6046511627906976e-05, |
|
"loss": 0.1212, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 8.744186046511627, |
|
"grad_norm": 0.7327284216880798, |
|
"learning_rate": 2.5116279069767445e-05, |
|
"loss": 0.0865, |
|
"step": 1880 |
|
}, |
|
{ |
|
"epoch": 8.790697674418604, |
|
"grad_norm": 2.343684673309326, |
|
"learning_rate": 2.4186046511627908e-05, |
|
"loss": 0.1467, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 8.837209302325581, |
|
"grad_norm": 5.241259574890137, |
|
"learning_rate": 2.3255813953488374e-05, |
|
"loss": 0.1658, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 8.837209302325581, |
|
"eval_accuracy": 0.9042904290429042, |
|
"eval_loss": 0.3840368390083313, |
|
"eval_runtime": 8.2921, |
|
"eval_samples_per_second": 73.081, |
|
"eval_steps_per_second": 9.165, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 8.883720930232558, |
|
"grad_norm": 0.7707066535949707, |
|
"learning_rate": 2.2325581395348837e-05, |
|
"loss": 0.1274, |
|
"step": 1910 |
|
}, |
|
{ |
|
"epoch": 8.930232558139535, |
|
"grad_norm": 2.1124603748321533, |
|
"learning_rate": 2.1395348837209303e-05, |
|
"loss": 0.1397, |
|
"step": 1920 |
|
}, |
|
{ |
|
"epoch": 8.976744186046512, |
|
"grad_norm": 0.4913456439971924, |
|
"learning_rate": 2.046511627906977e-05, |
|
"loss": 0.1697, |
|
"step": 1930 |
|
}, |
|
{ |
|
"epoch": 9.023255813953488, |
|
"grad_norm": 1.9249141216278076, |
|
"learning_rate": 1.9534883720930235e-05, |
|
"loss": 0.139, |
|
"step": 1940 |
|
}, |
|
{ |
|
"epoch": 9.069767441860465, |
|
"grad_norm": 1.6021732091903687, |
|
"learning_rate": 1.8604651162790697e-05, |
|
"loss": 0.1806, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 9.116279069767442, |
|
"grad_norm": 0.47562333941459656, |
|
"learning_rate": 1.7674418604651163e-05, |
|
"loss": 0.1272, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 9.162790697674419, |
|
"grad_norm": 4.387588024139404, |
|
"learning_rate": 1.674418604651163e-05, |
|
"loss": 0.1565, |
|
"step": 1970 |
|
}, |
|
{ |
|
"epoch": 9.209302325581396, |
|
"grad_norm": 0.03679787367582321, |
|
"learning_rate": 1.5813953488372095e-05, |
|
"loss": 0.039, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 9.255813953488373, |
|
"grad_norm": 0.04448888823390007, |
|
"learning_rate": 1.488372093023256e-05, |
|
"loss": 0.084, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 9.30232558139535, |
|
"grad_norm": 0.5206286311149597, |
|
"learning_rate": 1.3953488372093024e-05, |
|
"loss": 0.1, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 9.30232558139535, |
|
"eval_accuracy": 0.900990099009901, |
|
"eval_loss": 0.3872542083263397, |
|
"eval_runtime": 8.0752, |
|
"eval_samples_per_second": 75.044, |
|
"eval_steps_per_second": 9.411, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 9.348837209302326, |
|
"grad_norm": 0.08173186331987381, |
|
"learning_rate": 1.3023255813953488e-05, |
|
"loss": 0.1023, |
|
"step": 2010 |
|
}, |
|
{ |
|
"epoch": 9.395348837209303, |
|
"grad_norm": 0.45058727264404297, |
|
"learning_rate": 1.2093023255813954e-05, |
|
"loss": 0.1194, |
|
"step": 2020 |
|
}, |
|
{ |
|
"epoch": 9.44186046511628, |
|
"grad_norm": 0.05082235485315323, |
|
"learning_rate": 1.1162790697674418e-05, |
|
"loss": 0.0994, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 9.488372093023255, |
|
"grad_norm": 0.16433951258659363, |
|
"learning_rate": 1.0232558139534884e-05, |
|
"loss": 0.124, |
|
"step": 2040 |
|
}, |
|
{ |
|
"epoch": 9.534883720930232, |
|
"grad_norm": 3.3077304363250732, |
|
"learning_rate": 9.302325581395349e-06, |
|
"loss": 0.1311, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 9.581395348837209, |
|
"grad_norm": 3.5823140144348145, |
|
"learning_rate": 8.372093023255815e-06, |
|
"loss": 0.0295, |
|
"step": 2060 |
|
}, |
|
{ |
|
"epoch": 9.627906976744185, |
|
"grad_norm": 0.6963232159614563, |
|
"learning_rate": 7.44186046511628e-06, |
|
"loss": 0.0753, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 9.674418604651162, |
|
"grad_norm": 0.858066201210022, |
|
"learning_rate": 6.511627906976744e-06, |
|
"loss": 0.0945, |
|
"step": 2080 |
|
}, |
|
{ |
|
"epoch": 9.720930232558139, |
|
"grad_norm": 0.16633135080337524, |
|
"learning_rate": 5.581395348837209e-06, |
|
"loss": 0.1215, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 9.767441860465116, |
|
"grad_norm": 1.5643835067749023, |
|
"learning_rate": 4.651162790697674e-06, |
|
"loss": 0.1045, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 9.767441860465116, |
|
"eval_accuracy": 0.9075907590759076, |
|
"eval_loss": 0.37872469425201416, |
|
"eval_runtime": 8.0071, |
|
"eval_samples_per_second": 75.683, |
|
"eval_steps_per_second": 9.492, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 9.813953488372093, |
|
"grad_norm": 2.010878324508667, |
|
"learning_rate": 3.72093023255814e-06, |
|
"loss": 0.0887, |
|
"step": 2110 |
|
}, |
|
{ |
|
"epoch": 9.86046511627907, |
|
"grad_norm": 2.305905342102051, |
|
"learning_rate": 2.7906976744186046e-06, |
|
"loss": 0.1151, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 9.906976744186046, |
|
"grad_norm": 2.8573191165924072, |
|
"learning_rate": 1.86046511627907e-06, |
|
"loss": 0.1443, |
|
"step": 2130 |
|
}, |
|
{ |
|
"epoch": 9.953488372093023, |
|
"grad_norm": 3.0255908966064453, |
|
"learning_rate": 9.30232558139535e-07, |
|
"loss": 0.1765, |
|
"step": 2140 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"grad_norm": 2.159597635269165, |
|
"learning_rate": 0.0, |
|
"loss": 0.1705, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"step": 2150, |
|
"total_flos": 2.659620429501235e+18, |
|
"train_loss": 0.3051174351919529, |
|
"train_runtime": 1549.0078, |
|
"train_samples_per_second": 22.156, |
|
"train_steps_per_second": 1.388 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 2150, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 100, |
|
"total_flos": 2.659620429501235e+18, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|