| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.666666666666667, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 6.197302781887481, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.9987, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.5427567745628, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.045, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.789287104244284, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0106, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.8513067975000648, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9517, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 4.131500038693086, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9326, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.321074987447994, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9418, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.189499697165511, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9087, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 3.0655869330638934, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9025, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.3280983469185017, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8568, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.5445225991088278, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8093, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 2.363324099528786, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.2183, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.2019989157945463, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7372, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.3781140941279373, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7669, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.0254995803449134, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7786, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.9443210339747068, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.638, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.2711384284961342, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7749, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.892509163801348, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7035, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.938034208149548, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7336, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.884633074137525, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6761, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.6915082363916647, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6988, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1314241215972494, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0471, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.8257450610952127, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6647, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6394461687460596, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6442, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.8675267879165466, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6353, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.6202655683392411, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6423, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 1.1481883920998037, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6258, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.8468237228699831, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.618, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.8450161019220843, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6272, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.8758881389706432, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6238, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.6351367672727709, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6222, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.6586239486379113, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5992, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.8817671259591623, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.902, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.5821979898057683, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5824, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.5064216834780049, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5646, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.5406135509320869, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5887, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.47264625555385514, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.4934, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.5303319829243548, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5739, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.5000963645923708, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5631, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.5628145087838315, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5547, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.41356521082084535, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5086, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.5920164182683606, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.6039, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6373203146834688, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8322, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.4124742787522498, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5155, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.48575280640453916, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5281, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.4002413604909856, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5182, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.4264759038578856, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5123, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.45425587376579724, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5081, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.4363914465445056, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5091, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.4089816630801043, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5116, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.4292687733022561, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5067, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.3835314421416425, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5014, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.361905938152125, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5071, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.5468165234044171, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.7702, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.34001228831564034, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4656, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.3124033080167707, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4624, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.34496383226420974, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4809, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.33682315729010764, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4855, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.3038069519951789, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4591, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.3252658504420996, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.5317, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.293687896458208, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4763, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.30984816785419883, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4758, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.2987637742419163, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5037, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.4463426039999235, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7154, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.2813431746964439, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4831, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.26130601928727654, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.482, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.2934386011572668, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4852, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.26510379103343995, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.471, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.2615865305771671, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4646, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.2683548632717519, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4761, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.273865944169795, | |
| "learning_rate": 0.0, | |
| "loss": 0.4887, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 1.878724537357435e+17, | |
| "train_loss": 0.6499480456113815, | |
| "train_runtime": 3083.03, | |
| "train_samples_per_second": 2.27, | |
| "train_steps_per_second": 0.023 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 70, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.878724537357435e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |