| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.028985507246376812, | |
| "grad_norm": 11.274454167048727, | |
| "learning_rate": 0.0, | |
| "loss": 0.6357, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.057971014492753624, | |
| "grad_norm": 10.143348640958312, | |
| "learning_rate": 9.090909090909091e-07, | |
| "loss": 0.6666, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 9.086409087474683, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 0.6451, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11594202898550725, | |
| "grad_norm": 8.70666602748876, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.6048, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.14492753623188406, | |
| "grad_norm": 8.976482480167713, | |
| "learning_rate": 3.6363636363636366e-06, | |
| "loss": 0.5937, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 6.543196052759934, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.4943, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2028985507246377, | |
| "grad_norm": 7.511256155710179, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.4247, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2318840579710145, | |
| "grad_norm": 3.274069972289533, | |
| "learning_rate": 6.363636363636364e-06, | |
| "loss": 0.302, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 5.507527578193957, | |
| "learning_rate": 7.272727272727273e-06, | |
| "loss": 0.2751, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2898550724637681, | |
| "grad_norm": 3.651481150079746, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.2311, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3188405797101449, | |
| "grad_norm": 2.9570009811876314, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.2133, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 2.5680951348234355, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2021, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.37681159420289856, | |
| "grad_norm": 2.4726534615132394, | |
| "learning_rate": 9.997020702755353e-06, | |
| "loss": 0.2086, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.4057971014492754, | |
| "grad_norm": 2.0168121750509207, | |
| "learning_rate": 9.98808636150624e-06, | |
| "loss": 0.1844, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 1.988964439347895, | |
| "learning_rate": 9.973207623475964e-06, | |
| "loss": 0.1561, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.463768115942029, | |
| "grad_norm": 1.703404936865681, | |
| "learning_rate": 9.952402219937817e-06, | |
| "loss": 0.1561, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.4927536231884058, | |
| "grad_norm": 1.4224019007386586, | |
| "learning_rate": 9.925694945084369e-06, | |
| "loss": 0.1359, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 1.390079768168245, | |
| "learning_rate": 9.893117626479778e-06, | |
| "loss": 0.1422, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5507246376811594, | |
| "grad_norm": 1.363268530198362, | |
| "learning_rate": 9.854709087130261e-06, | |
| "loss": 0.1448, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.5797101449275363, | |
| "grad_norm": 1.2167507253689909, | |
| "learning_rate": 9.810515099218004e-06, | |
| "loss": 0.1276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 1.1574568817316992, | |
| "learning_rate": 9.76058832955357e-06, | |
| "loss": 0.1177, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6376811594202898, | |
| "grad_norm": 1.2101868179533353, | |
| "learning_rate": 9.704988276811883e-06, | |
| "loss": 0.1086, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 1.1144422829365617, | |
| "learning_rate": 9.643781200626512e-06, | |
| "loss": 0.1291, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 1.2450794816372637, | |
| "learning_rate": 9.577040042626832e-06, | |
| "loss": 0.1183, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7246376811594203, | |
| "grad_norm": 1.2167862630409976, | |
| "learning_rate": 9.504844339512096e-06, | |
| "loss": 0.1213, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7536231884057971, | |
| "grad_norm": 1.248557822116599, | |
| "learning_rate": 9.427280128266049e-06, | |
| "loss": 0.1177, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 1.113733087204506, | |
| "learning_rate": 9.344439843625034e-06, | |
| "loss": 0.1152, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.8115942028985508, | |
| "grad_norm": 1.1038118432089006, | |
| "learning_rate": 9.256422207921757e-06, | |
| "loss": 0.1246, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8405797101449275, | |
| "grad_norm": 1.100623510584588, | |
| "learning_rate": 9.163332113436031e-06, | |
| "loss": 0.1184, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 1.0179727735610475, | |
| "learning_rate": 9.065280497392663e-06, | |
| "loss": 0.1096, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8985507246376812, | |
| "grad_norm": 1.2701457238654967, | |
| "learning_rate": 8.962384209755453e-06, | |
| "loss": 0.1195, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 1.1294896292998675, | |
| "learning_rate": 8.854765873974898e-06, | |
| "loss": 0.1105, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 1.1171814840270677, | |
| "learning_rate": 8.742553740855507e-06, | |
| "loss": 0.1175, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9855072463768116, | |
| "grad_norm": 0.9886273247393406, | |
| "learning_rate": 8.625881535716883e-06, | |
| "loss": 0.1058, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.9886273247393406, | |
| "learning_rate": 8.504888299030748e-06, | |
| "loss": 0.1001, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0289855072463767, | |
| "grad_norm": 1.5562818130049305, | |
| "learning_rate": 8.379718220723772e-06, | |
| "loss": 0.0727, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0579710144927537, | |
| "grad_norm": 0.8147596643901716, | |
| "learning_rate": 8.250520468343722e-06, | |
| "loss": 0.0871, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 0.785533808227011, | |
| "learning_rate": 8.117449009293668e-06, | |
| "loss": 0.0752, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.1159420289855073, | |
| "grad_norm": 1.067197574454516, | |
| "learning_rate": 7.980662427346127e-06, | |
| "loss": 0.0862, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.144927536231884, | |
| "grad_norm": 1.0067934230612672, | |
| "learning_rate": 7.84032373365578e-06, | |
| "loss": 0.0928, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1739130434782608, | |
| "grad_norm": 1.027473082943366, | |
| "learning_rate": 7.696600172495997e-06, | |
| "loss": 0.0757, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.2028985507246377, | |
| "grad_norm": 0.9703955436663084, | |
| "learning_rate": 7.5496630219506805e-06, | |
| "loss": 0.0621, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.2318840579710144, | |
| "grad_norm": 0.9885265576714246, | |
| "learning_rate": 7.399687389798933e-06, | |
| "loss": 0.0855, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.2608695652173914, | |
| "grad_norm": 0.9758981949816004, | |
| "learning_rate": 7.246852004835807e-06, | |
| "loss": 0.084, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.289855072463768, | |
| "grad_norm": 0.9024594434416954, | |
| "learning_rate": 7.091339003877826e-06, | |
| "loss": 0.0767, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.318840579710145, | |
| "grad_norm": 0.830224350365886, | |
| "learning_rate": 6.933333714707094e-06, | |
| "loss": 0.0806, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.3478260869565217, | |
| "grad_norm": 0.8475268092069895, | |
| "learning_rate": 6.773024435212678e-06, | |
| "loss": 0.0686, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.3768115942028984, | |
| "grad_norm": 0.8152669559501473, | |
| "learning_rate": 6.6106022089924535e-06, | |
| "loss": 0.0913, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.4057971014492754, | |
| "grad_norm": 0.8455140548788239, | |
| "learning_rate": 6.4462605976828395e-06, | |
| "loss": 0.0746, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.434782608695652, | |
| "grad_norm": 0.817492348372533, | |
| "learning_rate": 6.280195450287736e-06, | |
| "loss": 0.0976, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.463768115942029, | |
| "grad_norm": 0.880952546066598, | |
| "learning_rate": 6.112604669781572e-06, | |
| "loss": 0.0833, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4927536231884058, | |
| "grad_norm": 0.8489601481147181, | |
| "learning_rate": 5.943687977264584e-06, | |
| "loss": 0.0784, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 0.8142122839258903, | |
| "learning_rate": 5.773646673951406e-06, | |
| "loss": 0.0838, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.5507246376811594, | |
| "grad_norm": 0.8922929202950094, | |
| "learning_rate": 5.6026834012766155e-06, | |
| "loss": 0.0828, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5797101449275361, | |
| "grad_norm": 0.7122624377371234, | |
| "learning_rate": 5.4310018994030974e-06, | |
| "loss": 0.0797, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.608695652173913, | |
| "grad_norm": 0.9217981197726534, | |
| "learning_rate": 5.258806764421048e-06, | |
| "loss": 0.0772, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.6376811594202898, | |
| "grad_norm": 0.9257391753751778, | |
| "learning_rate": 5.0863032045269435e-06, | |
| "loss": 0.0766, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.8307688279849595, | |
| "learning_rate": 4.913696795473058e-06, | |
| "loss": 0.0818, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.6956521739130435, | |
| "grad_norm": 0.9173954819802914, | |
| "learning_rate": 4.741193235578953e-06, | |
| "loss": 0.0927, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.7246376811594204, | |
| "grad_norm": 0.8708594385417067, | |
| "learning_rate": 4.568998100596903e-06, | |
| "loss": 0.0665, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7536231884057971, | |
| "grad_norm": 0.8126974465329198, | |
| "learning_rate": 4.397316598723385e-06, | |
| "loss": 0.0839, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.7826086956521738, | |
| "grad_norm": 0.853158086441921, | |
| "learning_rate": 4.226353326048594e-06, | |
| "loss": 0.0773, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.8115942028985508, | |
| "grad_norm": 0.8525307891771787, | |
| "learning_rate": 4.056312022735417e-06, | |
| "loss": 0.0809, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.8405797101449275, | |
| "grad_norm": 1.0467111292284437, | |
| "learning_rate": 3.887395330218429e-06, | |
| "loss": 0.0782, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.8695652173913042, | |
| "grad_norm": 0.9286296034389645, | |
| "learning_rate": 3.7198045497122647e-06, | |
| "loss": 0.0774, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.8985507246376812, | |
| "grad_norm": 0.7656482475139829, | |
| "learning_rate": 3.553739402317162e-06, | |
| "loss": 0.0767, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.927536231884058, | |
| "grad_norm": 0.7766347569440668, | |
| "learning_rate": 3.389397791007548e-06, | |
| "loss": 0.0984, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 0.7336465690307817, | |
| "learning_rate": 3.226975564787322e-06, | |
| "loss": 0.074, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.9855072463768115, | |
| "grad_norm": 0.795203124951574, | |
| "learning_rate": 3.0666662852929063e-06, | |
| "loss": 0.0785, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.2688010574239046, | |
| "learning_rate": 2.9086609961221758e-06, | |
| "loss": 0.0659, | |
| "step": 70 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 102, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6977636007936.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |