| { | |
| "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": 1.2167548650541202, | |
| "learning_rate": 9.810515099218004e-06, | |
| "loss": 0.1276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 1.2149476390607494, | |
| "learning_rate": 9.76058832955357e-06, | |
| "loss": 0.118, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6376811594202898, | |
| "grad_norm": 1.1518569512633139, | |
| "learning_rate": 9.704988276811883e-06, | |
| "loss": 0.1088, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 1.1392991979040483, | |
| "learning_rate": 9.643781200626512e-06, | |
| "loss": 0.1287, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 1.2433820634735937, | |
| "learning_rate": 9.577040042626832e-06, | |
| "loss": 0.1186, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7246376811594203, | |
| "grad_norm": 1.230070806896417, | |
| "learning_rate": 9.504844339512096e-06, | |
| "loss": 0.1212, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7536231884057971, | |
| "grad_norm": 1.276827967155928, | |
| "learning_rate": 9.427280128266049e-06, | |
| "loss": 0.1173, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 1.1539744752759111, | |
| "learning_rate": 9.344439843625034e-06, | |
| "loss": 0.1145, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.8115942028985508, | |
| "grad_norm": 1.101305700093271, | |
| "learning_rate": 9.256422207921757e-06, | |
| "loss": 0.1246, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8405797101449275, | |
| "grad_norm": 1.0720557642935318, | |
| "learning_rate": 9.163332113436031e-06, | |
| "loss": 0.118, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 1.0709210914227136, | |
| "learning_rate": 9.065280497392663e-06, | |
| "loss": 0.1097, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8985507246376812, | |
| "grad_norm": 1.3387140042516312, | |
| "learning_rate": 8.962384209755453e-06, | |
| "loss": 0.119, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 1.2000636204494934, | |
| "learning_rate": 8.854765873974898e-06, | |
| "loss": 0.1103, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 1.1660635680582712, | |
| "learning_rate": 8.742553740855507e-06, | |
| "loss": 0.1168, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9855072463768116, | |
| "grad_norm": 1.0295558925452102, | |
| "learning_rate": 8.625881535716883e-06, | |
| "loss": 0.1047, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.0295558925452102, | |
| "learning_rate": 8.504888299030748e-06, | |
| "loss": 0.0989, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0289855072463767, | |
| "grad_norm": 1.506591807470232, | |
| "learning_rate": 8.379718220723772e-06, | |
| "loss": 0.0736, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0579710144927537, | |
| "grad_norm": 0.846102643595135, | |
| "learning_rate": 8.250520468343722e-06, | |
| "loss": 0.0876, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 0.8226526583596483, | |
| "learning_rate": 8.117449009293668e-06, | |
| "loss": 0.0773, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.1159420289855073, | |
| "grad_norm": 1.0811685151313555, | |
| "learning_rate": 7.980662427346127e-06, | |
| "loss": 0.0855, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.144927536231884, | |
| "grad_norm": 0.9305493105996977, | |
| "learning_rate": 7.84032373365578e-06, | |
| "loss": 0.0923, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1739130434782608, | |
| "grad_norm": 0.9787521121239098, | |
| "learning_rate": 7.696600172495997e-06, | |
| "loss": 0.0747, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.2028985507246377, | |
| "grad_norm": 0.9440280872408277, | |
| "learning_rate": 7.5496630219506805e-06, | |
| "loss": 0.0624, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.2318840579710144, | |
| "grad_norm": 1.0168023667319657, | |
| "learning_rate": 7.399687389798933e-06, | |
| "loss": 0.0862, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.2608695652173914, | |
| "grad_norm": 1.0534520932326086, | |
| "learning_rate": 7.246852004835807e-06, | |
| "loss": 0.0832, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.289855072463768, | |
| "grad_norm": 0.8939408658703963, | |
| "learning_rate": 7.091339003877826e-06, | |
| "loss": 0.0779, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.318840579710145, | |
| "grad_norm": 0.8502452526837244, | |
| "learning_rate": 6.933333714707094e-06, | |
| "loss": 0.0808, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.3478260869565217, | |
| "grad_norm": 0.838670630249861, | |
| "learning_rate": 6.773024435212678e-06, | |
| "loss": 0.0688, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.3768115942028984, | |
| "grad_norm": 0.801607798732365, | |
| "learning_rate": 6.6106022089924535e-06, | |
| "loss": 0.0897, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.4057971014492754, | |
| "grad_norm": 0.8006660956713604, | |
| "learning_rate": 6.4462605976828395e-06, | |
| "loss": 0.0742, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.434782608695652, | |
| "grad_norm": 0.8042569883275191, | |
| "learning_rate": 6.280195450287736e-06, | |
| "loss": 0.0953, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.463768115942029, | |
| "grad_norm": 0.9038324935270676, | |
| "learning_rate": 6.112604669781572e-06, | |
| "loss": 0.0839, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4927536231884058, | |
| "grad_norm": 0.8460692397365317, | |
| "learning_rate": 5.943687977264584e-06, | |
| "loss": 0.0771, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 0.8300524398027148, | |
| "learning_rate": 5.773646673951406e-06, | |
| "loss": 0.0824, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.5507246376811594, | |
| "grad_norm": 0.8512930075204654, | |
| "learning_rate": 5.6026834012766155e-06, | |
| "loss": 0.0835, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5797101449275361, | |
| "grad_norm": 0.7732172809054101, | |
| "learning_rate": 5.4310018994030974e-06, | |
| "loss": 0.0787, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.608695652173913, | |
| "grad_norm": 0.8694598913019425, | |
| "learning_rate": 5.258806764421048e-06, | |
| "loss": 0.0764, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.6376811594202898, | |
| "grad_norm": 0.9361374054111669, | |
| "learning_rate": 5.0863032045269435e-06, | |
| "loss": 0.0773, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.8070286203847241, | |
| "learning_rate": 4.913696795473058e-06, | |
| "loss": 0.0814, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.6956521739130435, | |
| "grad_norm": 0.9276738248259214, | |
| "learning_rate": 4.741193235578953e-06, | |
| "loss": 0.0924, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.7246376811594204, | |
| "grad_norm": 0.8188739154958693, | |
| "learning_rate": 4.568998100596903e-06, | |
| "loss": 0.0659, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7536231884057971, | |
| "grad_norm": 0.7835730541507403, | |
| "learning_rate": 4.397316598723385e-06, | |
| "loss": 0.0827, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.7826086956521738, | |
| "grad_norm": 0.8254420427828969, | |
| "learning_rate": 4.226353326048594e-06, | |
| "loss": 0.0777, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.8115942028985508, | |
| "grad_norm": 0.8014760946487198, | |
| "learning_rate": 4.056312022735417e-06, | |
| "loss": 0.0799, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.8405797101449275, | |
| "grad_norm": 0.909284799411466, | |
| "learning_rate": 3.887395330218429e-06, | |
| "loss": 0.0769, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.8695652173913042, | |
| "grad_norm": 0.9101996746678516, | |
| "learning_rate": 3.7198045497122647e-06, | |
| "loss": 0.0783, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.8985507246376812, | |
| "grad_norm": 0.7981223504922568, | |
| "learning_rate": 3.553739402317162e-06, | |
| "loss": 0.0769, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.927536231884058, | |
| "grad_norm": 0.8131851608148943, | |
| "learning_rate": 3.389397791007548e-06, | |
| "loss": 0.0975, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 0.7422094984620301, | |
| "learning_rate": 3.226975564787322e-06, | |
| "loss": 0.0755, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.9855072463768115, | |
| "grad_norm": 0.8122097222072056, | |
| "learning_rate": 3.0666662852929063e-06, | |
| "loss": 0.079, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.2716537005367277, | |
| "learning_rate": 2.9086609961221758e-06, | |
| "loss": 0.066, | |
| "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 | |
| } | |