| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.009146306607063235, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001829261321412647, | |
| "grad_norm": 2.0105717182159424, | |
| "learning_rate": 5e-05, | |
| "loss": 3.631, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001829261321412647, | |
| "eval_loss": 6.472896575927734, | |
| "eval_runtime": 3.4432, | |
| "eval_samples_per_second": 14.521, | |
| "eval_steps_per_second": 3.776, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003658522642825294, | |
| "grad_norm": 1.8902631998062134, | |
| "learning_rate": 0.0001, | |
| "loss": 4.4256, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0005487783964237942, | |
| "grad_norm": 1.8536314964294434, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 4.7617, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0007317045285650588, | |
| "grad_norm": 1.9695987701416016, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 4.9071, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0009146306607063236, | |
| "grad_norm": 2.246405601501465, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 4.9707, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0010975567928475883, | |
| "grad_norm": 2.866654634475708, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 5.0945, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001280482924988853, | |
| "grad_norm": 3.293781042098999, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 5.0133, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0014634090571301176, | |
| "grad_norm": 3.599553346633911, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 4.8596, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0016463351892713823, | |
| "grad_norm": 3.323606252670288, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 4.6907, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0018292613214126471, | |
| "grad_norm": 3.172093629837036, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 4.7782, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002012187453553912, | |
| "grad_norm": 3.901989459991455, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 4.3137, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0021951135856951767, | |
| "grad_norm": 4.4987030029296875, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 4.1896, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002378039717836441, | |
| "grad_norm": 5.5157647132873535, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 3.5969, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.002560965849977706, | |
| "grad_norm": 5.361765384674072, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 3.6721, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0027438919821189704, | |
| "grad_norm": 3.710472345352173, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 3.672, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0029268181142602353, | |
| "grad_norm": 3.0013794898986816, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 3.7136, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0031097442464015, | |
| "grad_norm": 2.4556336402893066, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 3.6958, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0032926703785427646, | |
| "grad_norm": 1.9984523057937622, | |
| "learning_rate": 7.75e-05, | |
| "loss": 3.5622, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0034755965106840294, | |
| "grad_norm": 1.4705381393432617, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 3.6139, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0036585226428252943, | |
| "grad_norm": 1.416642427444458, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 3.5838, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0038414487749665587, | |
| "grad_norm": 1.7146713733673096, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 3.7138, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004024374907107824, | |
| "grad_norm": 2.052887201309204, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 3.7816, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0042073010392490885, | |
| "grad_norm": 2.7639307975769043, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 3.7285, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.004390227171390353, | |
| "grad_norm": 2.4564733505249023, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 3.6091, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.004573153303531617, | |
| "grad_norm": 4.71488094329834, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 4.3749, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.004573153303531617, | |
| "eval_loss": 3.553548574447632, | |
| "eval_runtime": 3.4893, | |
| "eval_samples_per_second": 14.329, | |
| "eval_steps_per_second": 3.726, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.004756079435672882, | |
| "grad_norm": 1.7480310201644897, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 3.017, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.004939005567814147, | |
| "grad_norm": 1.3907936811447144, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 3.1465, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.005121931699955412, | |
| "grad_norm": 26.612220764160156, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 3.284, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.005304857832096677, | |
| "grad_norm": 1.115344762802124, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 3.3917, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.005487783964237941, | |
| "grad_norm": 1.0461210012435913, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 3.4146, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005670710096379206, | |
| "grad_norm": 1.325683355331421, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 3.4329, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0058536362285204705, | |
| "grad_norm": 1.1475805044174194, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 3.4794, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.006036562360661735, | |
| "grad_norm": 1.2684738636016846, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 3.5487, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.006219488492803, | |
| "grad_norm": 1.2360631227493286, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 3.6217, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.006402414624944265, | |
| "grad_norm": 1.497918725013733, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 3.5036, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.006585340757085529, | |
| "grad_norm": 1.5676883459091187, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 3.5703, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006768266889226794, | |
| "grad_norm": 2.1435370445251465, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 3.6886, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.006951193021368059, | |
| "grad_norm": 1.654815435409546, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 3.0422, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.007134119153509324, | |
| "grad_norm": 1.5040534734725952, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 3.1364, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.007317045285650589, | |
| "grad_norm": 1.3657926321029663, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 3.1935, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.007499971417791853, | |
| "grad_norm": 1.264070987701416, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 3.4067, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0076828975499331174, | |
| "grad_norm": 1.3875404596328735, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 3.2835, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.007865823682074382, | |
| "grad_norm": 1.2422032356262207, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 3.4278, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.008048749814215647, | |
| "grad_norm": 1.25910222530365, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 3.398, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.008231675946356912, | |
| "grad_norm": 1.284754753112793, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 3.5085, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008414602078498177, | |
| "grad_norm": 1.3367297649383545, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 3.4191, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.008597528210639442, | |
| "grad_norm": 1.2780907154083252, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 3.3408, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.008780454342780707, | |
| "grad_norm": 1.462539553642273, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 3.653, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.00896338047492197, | |
| "grad_norm": 1.7677708864212036, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 3.4973, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.009146306607063235, | |
| "grad_norm": 3.7667927742004395, | |
| "learning_rate": 1e-05, | |
| "loss": 3.8798, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.009146306607063235, | |
| "eval_loss": 3.431530237197876, | |
| "eval_runtime": 3.5348, | |
| "eval_samples_per_second": 14.145, | |
| "eval_steps_per_second": 3.678, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.968083617316864e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |