| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03961867029837811, | |
| "eval_steps": 20, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006603111716396351, | |
| "grad_norm": 300608.875, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 118.0917, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006603111716396351, | |
| "eval_loss": 15.74350357055664, | |
| "eval_runtime": 6.9487, | |
| "eval_samples_per_second": 71.236, | |
| "eval_steps_per_second": 71.236, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013206223432792703, | |
| "grad_norm": 277566.0, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 120.2507, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0019809335149189055, | |
| "grad_norm": 418405.1875, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 143.5985, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0026412446865585405, | |
| "grad_norm": 222641.09375, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 113.3078, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003301555858198176, | |
| "grad_norm": 80659.1640625, | |
| "learning_rate": 0.00015, | |
| "loss": 89.0113, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003961867029837811, | |
| "grad_norm": 174515.8125, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 92.8388, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004622178201477446, | |
| "grad_norm": 68650.6640625, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 88.8969, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005282489373117081, | |
| "grad_norm": 52668.09765625, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 83.2267, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005942800544756717, | |
| "grad_norm": 61407.41796875, | |
| "learning_rate": 0.00027, | |
| "loss": 92.7579, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006603111716396352, | |
| "grad_norm": 37122.61328125, | |
| "learning_rate": 0.0003, | |
| "loss": 84.5321, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007263422888035987, | |
| "grad_norm": 58673.69921875, | |
| "learning_rate": 0.0002999911984174669, | |
| "loss": 93.7724, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007923734059675622, | |
| "grad_norm": 121362.359375, | |
| "learning_rate": 0.0002999647947027726, | |
| "loss": 108.6176, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008584045231315257, | |
| "grad_norm": 233485.53125, | |
| "learning_rate": 0.0002999207919545099, | |
| "loss": 105.0768, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009244356402954892, | |
| "grad_norm": 318686.03125, | |
| "learning_rate": 0.0002998591953365965, | |
| "loss": 99.2428, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009904667574594527, | |
| "grad_norm": 101194.0625, | |
| "learning_rate": 0.00029978001207766854, | |
| "loss": 95.0131, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010564978746234162, | |
| "grad_norm": 83839.953125, | |
| "learning_rate": 0.00029968325147023263, | |
| "loss": 86.1607, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.011225289917873797, | |
| "grad_norm": 68342.7109375, | |
| "learning_rate": 0.000299568924869575, | |
| "loss": 85.744, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011885601089513434, | |
| "grad_norm": 232218.078125, | |
| "learning_rate": 0.00029943704569242917, | |
| "loss": 89.0831, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012545912261153069, | |
| "grad_norm": 142226.359375, | |
| "learning_rate": 0.0002992876294154013, | |
| "loss": 87.8575, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.013206223432792704, | |
| "grad_norm": 99313.8125, | |
| "learning_rate": 0.00029912069357315393, | |
| "loss": 91.0717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013206223432792704, | |
| "eval_loss": 15.472484588623047, | |
| "eval_runtime": 6.5506, | |
| "eval_samples_per_second": 75.565, | |
| "eval_steps_per_second": 75.565, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013866534604432339, | |
| "grad_norm": 361644.25, | |
| "learning_rate": 0.00029893625775634835, | |
| "loss": 97.0436, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014526845776071974, | |
| "grad_norm": 1564964.125, | |
| "learning_rate": 0.0002987343436093454, | |
| "loss": 279.3873, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.015187156947711609, | |
| "grad_norm": 1539204.875, | |
| "learning_rate": 0.00029851497482766547, | |
| "loss": 668.731, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.015847468119351244, | |
| "grad_norm": 2246419.0, | |
| "learning_rate": 0.00029827817715520773, | |
| "loss": 758.2188, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01650777929099088, | |
| "grad_norm": 2776799.5, | |
| "learning_rate": 0.0002980239783812289, | |
| "loss": 546.4688, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.017168090462630514, | |
| "grad_norm": 1617533.25, | |
| "learning_rate": 0.0002977524083370822, | |
| "loss": 422.7344, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01782840163427015, | |
| "grad_norm": 450934.75, | |
| "learning_rate": 0.00029746349889271645, | |
| "loss": 339.3945, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.018488712805909784, | |
| "grad_norm": 2286499.5, | |
| "learning_rate": 0.0002971572839529358, | |
| "loss": 236.2812, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01914902397754942, | |
| "grad_norm": 3999992.25, | |
| "learning_rate": 0.00029683379945342125, | |
| "loss": 159.8123, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.019809335149189054, | |
| "grad_norm": 308621.125, | |
| "learning_rate": 0.000296493083356513, | |
| "loss": 124.3508, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02046964632082869, | |
| "grad_norm": 497120.59375, | |
| "learning_rate": 0.00029613517564675565, | |
| "loss": 138.8941, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.021129957492468324, | |
| "grad_norm": 99928.7890625, | |
| "learning_rate": 0.0002957601183262058, | |
| "loss": 151.8223, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02179026866410796, | |
| "grad_norm": 405539.0625, | |
| "learning_rate": 0.000295367955409503, | |
| "loss": 1203.2074, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.022450579835747594, | |
| "grad_norm": 463994.21875, | |
| "learning_rate": 0.00029495873291870436, | |
| "loss": 1256.5367, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02311089100738723, | |
| "grad_norm": 953879.5625, | |
| "learning_rate": 0.0002945324988778834, | |
| "loss": 938.6675, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.023771202179026868, | |
| "grad_norm": 178221.28125, | |
| "learning_rate": 0.00029408930330749477, | |
| "loss": 356.9532, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0244315133506665, | |
| "grad_norm": 146182.625, | |
| "learning_rate": 0.0002936291982185036, | |
| "loss": 238.3703, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.025091824522306138, | |
| "grad_norm": 116065.7421875, | |
| "learning_rate": 0.00029315223760628217, | |
| "loss": 212.9555, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.02575213569394577, | |
| "grad_norm": 129193.03125, | |
| "learning_rate": 0.00029265847744427303, | |
| "loss": 227.929, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.026412446865585408, | |
| "grad_norm": 152996.75, | |
| "learning_rate": 0.00029214797567742035, | |
| "loss": 220.4361, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.026412446865585408, | |
| "eval_loss": 16.9798526763916, | |
| "eval_runtime": 6.5479, | |
| "eval_samples_per_second": 75.597, | |
| "eval_steps_per_second": 75.597, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02707275803722504, | |
| "grad_norm": 166313.5, | |
| "learning_rate": 0.00029162079221537, | |
| "loss": 178.7949, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.027733069208864678, | |
| "grad_norm": 135891.28125, | |
| "learning_rate": 0.0002910769889254386, | |
| "loss": 201.0785, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.02839338038050431, | |
| "grad_norm": 64060.7890625, | |
| "learning_rate": 0.0002905166296253533, | |
| "loss": 163.5094, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.029053691552143948, | |
| "grad_norm": 96362.4921875, | |
| "learning_rate": 0.0002899397800757626, | |
| "loss": 140.6384, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02971400272378358, | |
| "grad_norm": 166254.203125, | |
| "learning_rate": 0.0002893465079725187, | |
| "loss": 139.1684, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.030374313895423218, | |
| "grad_norm": 161925.5625, | |
| "learning_rate": 0.0002887368829387333, | |
| "loss": 140.9152, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.031034625067062855, | |
| "grad_norm": 637966.3125, | |
| "learning_rate": 0.0002881109765166071, | |
| "loss": 131.3419, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03169493623870249, | |
| "grad_norm": 367775.03125, | |
| "learning_rate": 0.00028746886215903387, | |
| "loss": 155.0525, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.032355247410342125, | |
| "grad_norm": 307120.84375, | |
| "learning_rate": 0.00028681061522098047, | |
| "loss": 148.0313, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03301555858198176, | |
| "grad_norm": 164428.203125, | |
| "learning_rate": 0.0002861363129506435, | |
| "loss": 139.0605, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03367586975362139, | |
| "grad_norm": 56655.140625, | |
| "learning_rate": 0.0002854460344803842, | |
| "loss": 105.2498, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.03433618092526103, | |
| "grad_norm": 110095.71875, | |
| "learning_rate": 0.00028473986081744163, | |
| "loss": 107.1039, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.034996492096900665, | |
| "grad_norm": 84727.8125, | |
| "learning_rate": 0.000284017874834426, | |
| "loss": 114.7597, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0356568032685403, | |
| "grad_norm": 108558.9140625, | |
| "learning_rate": 0.0002832801612595937, | |
| "loss": 131.0451, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03631711444017993, | |
| "grad_norm": 35595.47265625, | |
| "learning_rate": 0.0002825268066669034, | |
| "loss": 135.1516, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.03697742561181957, | |
| "grad_norm": 54421.08203125, | |
| "learning_rate": 0.00028175789946585693, | |
| "loss": 116.2731, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.037637736783459205, | |
| "grad_norm": 72844.515625, | |
| "learning_rate": 0.0002809735298911234, | |
| "loss": 101.419, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03829804795509884, | |
| "grad_norm": 58473.41015625, | |
| "learning_rate": 0.00028017378999195015, | |
| "loss": 101.8432, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.03895835912673848, | |
| "grad_norm": 41094.68359375, | |
| "learning_rate": 0.0002793587736213603, | |
| "loss": 114.5148, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.03961867029837811, | |
| "grad_norm": 75345.5234375, | |
| "learning_rate": 0.00027852857642513836, | |
| "loss": 119.3659, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03961867029837811, | |
| "eval_loss": 16.175268173217773, | |
| "eval_runtime": 6.5722, | |
| "eval_samples_per_second": 75.317, | |
| "eval_steps_per_second": 75.317, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 40390914736128.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |