| { | |
| "best_metric": 0.6870357394218445, | |
| "best_model_checkpoint": "../artifacts/LlaMa3-QLoRA-PatentMatch-v0.1/checkpoint-80", | |
| "epoch": 0.7547169811320755, | |
| "eval_steps": 20, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009433962264150943, | |
| "grad_norm": 20.337095260620117, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5913, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018867924528301886, | |
| "grad_norm": 11.506393432617188, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6775, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02830188679245283, | |
| "grad_norm": 9.703904151916504, | |
| "learning_rate": 2e-05, | |
| "loss": 0.576, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03773584905660377, | |
| "grad_norm": 11.118324279785156, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5084, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04716981132075472, | |
| "grad_norm": 13.329315185546875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5712, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05660377358490566, | |
| "grad_norm": 29.63173484802246, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7234, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0660377358490566, | |
| "grad_norm": 17.787134170532227, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6053, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07547169811320754, | |
| "grad_norm": 30.639591217041016, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6873, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08490566037735849, | |
| "grad_norm": 9.612072944641113, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5541, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09433962264150944, | |
| "grad_norm": 8.989519119262695, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7858, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10377358490566038, | |
| "grad_norm": 17.486469268798828, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9176, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11320754716981132, | |
| "grad_norm": 35.29791259765625, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6558, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12264150943396226, | |
| "grad_norm": 19.468692779541016, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6085, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1320754716981132, | |
| "grad_norm": 9.410886764526367, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6229, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.14150943396226415, | |
| "grad_norm": 11.87700080871582, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5764, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": 20.188251495361328, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8275, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.16037735849056603, | |
| "grad_norm": 28.298933029174805, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5896, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.16981132075471697, | |
| "grad_norm": 47.8366813659668, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8496, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1792452830188679, | |
| "grad_norm": 36.19501495361328, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6756, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18867924528301888, | |
| "grad_norm": 22.574682235717773, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6366, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18867924528301888, | |
| "eval_loss": 0.7738199234008789, | |
| "eval_runtime": 18.5622, | |
| "eval_samples_per_second": 15.893, | |
| "eval_steps_per_second": 3.179, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19811320754716982, | |
| "grad_norm": 47.26109313964844, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8049, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20754716981132076, | |
| "grad_norm": 14.115569114685059, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6604, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2169811320754717, | |
| "grad_norm": 25.182506561279297, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7591, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.22641509433962265, | |
| "grad_norm": 11.066629409790039, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6497, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2358490566037736, | |
| "grad_norm": 8.666443824768066, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5788, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24528301886792453, | |
| "grad_norm": 7.663419723510742, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7128, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.25471698113207547, | |
| "grad_norm": 30.738019943237305, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7349, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2641509433962264, | |
| "grad_norm": 29.7031307220459, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7618, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.27358490566037735, | |
| "grad_norm": 36.29247283935547, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6923, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.2830188679245283, | |
| "grad_norm": 16.721107482910156, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5942, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.29245283018867924, | |
| "grad_norm": 36.51066970825195, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7745, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3018867924528302, | |
| "grad_norm": 13.144597053527832, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6199, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3113207547169811, | |
| "grad_norm": 24.113306045532227, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6653, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.32075471698113206, | |
| "grad_norm": 34.57608413696289, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5586, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.330188679245283, | |
| "grad_norm": 15.308676719665527, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7438, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.33962264150943394, | |
| "grad_norm": 34.94574737548828, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7437, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3490566037735849, | |
| "grad_norm": 53.19334030151367, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8349, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3584905660377358, | |
| "grad_norm": 38.979618072509766, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6599, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.36792452830188677, | |
| "grad_norm": 30.653545379638672, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6782, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "grad_norm": 28.044891357421875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7945, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "eval_loss": 0.7332659959793091, | |
| "eval_runtime": 19.1011, | |
| "eval_samples_per_second": 15.444, | |
| "eval_steps_per_second": 3.089, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3867924528301887, | |
| "grad_norm": 7.029095649719238, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6402, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.39622641509433965, | |
| "grad_norm": 31.614521026611328, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6392, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4056603773584906, | |
| "grad_norm": 8.320149421691895, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5229, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.41509433962264153, | |
| "grad_norm": 18.34058380126953, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6935, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.42452830188679247, | |
| "grad_norm": 36.57161331176758, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6399, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4339622641509434, | |
| "grad_norm": 7.638645172119141, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7015, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.44339622641509435, | |
| "grad_norm": 18.424884796142578, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7157, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "grad_norm": 51.02284240722656, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8159, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.46226415094339623, | |
| "grad_norm": 29.55755615234375, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6529, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.4716981132075472, | |
| "grad_norm": 12.764640808105469, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6704, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4811320754716981, | |
| "grad_norm": 17.65540313720703, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6762, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.49056603773584906, | |
| "grad_norm": 10.487552642822266, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6094, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 10.158540725708008, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6539, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.5094339622641509, | |
| "grad_norm": 27.807415008544922, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7554, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5188679245283019, | |
| "grad_norm": 39.26100540161133, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8584, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5283018867924528, | |
| "grad_norm": 8.890057563781738, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7872, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.5377358490566038, | |
| "grad_norm": 11.212479591369629, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8501, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.5471698113207547, | |
| "grad_norm": 8.871652603149414, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6034, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5566037735849056, | |
| "grad_norm": 13.393775939941406, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5953, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.5660377358490566, | |
| "grad_norm": 16.56597328186035, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5595, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5660377358490566, | |
| "eval_loss": 0.7103222012519836, | |
| "eval_runtime": 19.0536, | |
| "eval_samples_per_second": 15.483, | |
| "eval_steps_per_second": 3.097, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5754716981132075, | |
| "grad_norm": 66.63365936279297, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8609, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.5849056603773585, | |
| "grad_norm": 43.89859390258789, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7555, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.5943396226415094, | |
| "grad_norm": 54.232025146484375, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7666, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6037735849056604, | |
| "grad_norm": 10.439966201782227, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6019, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.6132075471698113, | |
| "grad_norm": 15.057198524475098, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6797, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.6226415094339622, | |
| "grad_norm": 8.816701889038086, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8066, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6320754716981132, | |
| "grad_norm": 16.436609268188477, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5891, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.6415094339622641, | |
| "grad_norm": 27.5755672454834, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6204, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.6509433962264151, | |
| "grad_norm": 26.33946990966797, | |
| "learning_rate": 2e-05, | |
| "loss": 0.671, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.660377358490566, | |
| "grad_norm": 64.1870346069336, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7638, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6698113207547169, | |
| "grad_norm": 21.89188003540039, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6771, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.6792452830188679, | |
| "grad_norm": 8.088455200195312, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6761, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6886792452830188, | |
| "grad_norm": 11.988521575927734, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6315, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.6981132075471698, | |
| "grad_norm": 8.751002311706543, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5967, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.7075471698113207, | |
| "grad_norm": 22.44446563720703, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5986, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7169811320754716, | |
| "grad_norm": 6.895334243774414, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6324, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.7264150943396226, | |
| "grad_norm": 8.335739135742188, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6581, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.7358490566037735, | |
| "grad_norm": 6.27984619140625, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6899, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.7452830188679245, | |
| "grad_norm": 13.635252952575684, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7032, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "grad_norm": 5.515637397766113, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5121, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7547169811320755, | |
| "eval_loss": 0.6870357394218445, | |
| "eval_runtime": 19.1439, | |
| "eval_samples_per_second": 15.41, | |
| "eval_steps_per_second": 3.082, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 106, | |
| "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": 3.081598178082816e+16, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |