| { | |
| "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": 6.5314817126396605, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.039, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 6.411226211105682, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.007, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 6.03133238655822, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0241, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.7614466505709547, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9256, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 4.229667625274101, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9246, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.271242611972377, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9244, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 4.175033590443489, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9028, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.987240488063728, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8658, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.2208403898960354, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8271, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.6368209966441267, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8045, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 1.7407139578415196, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 1.1541, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.4903732825489864, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.8183, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.1387314756946674, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7249, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.9622986586737649, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.6702, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.202437985998727, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.697, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.055134927131131, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7302, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.8266818011748471, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6992, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.0098700924964996, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.6992, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.9110314401306685, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7385, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.7293179813266315, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6771, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4116554460831903, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.036, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.6717927696411902, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6562, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.7856660243507388, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6406, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.6907398105902497, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6382, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.578023955651648, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6324, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.7057408759686685, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.626, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.6695515982268823, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6271, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5483892523701176, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6146, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.5791212339845703, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6008, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.6227576341749919, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6045, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.56328099334905, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5922, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.917071302944978, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.8831, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.47392367492469706, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5957, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.6220069709358489, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5322, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.44117649459601027, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.574, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.5000691267266321, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5403, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.5238095238095237, | |
| "grad_norm": 0.5471569488938381, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5499, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 0.5206224380615669, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5704, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.4992983089213398, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5863, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.4532435197061412, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5153, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.9047619047619047, | |
| "grad_norm": 0.47188984858284133, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5561, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.7393169355556092, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.8115, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.095238095238095, | |
| "grad_norm": 0.4300138104002246, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5274, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.5486304902107627, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5042, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.3912659614188877, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.514, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.43859703727535104, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5102, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.476190476190476, | |
| "grad_norm": 0.4152072796435691, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5084, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.378279031747347, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5069, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.40105278834779906, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.4998, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.38485178131265124, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5121, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.3568969456024383, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5037, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.3917108399606805, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5084, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.0476190476190474, | |
| "grad_norm": 0.5625678926141774, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.7461, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.33638019603674757, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4874, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.41949833161994055, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4867, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.34434070204569706, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4773, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.31702096619813974, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4981, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.3100134818597013, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4494, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.619047619047619, | |
| "grad_norm": 0.6463917980073877, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.5017, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.3288971269174857, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4787, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.809523809523809, | |
| "grad_norm": 0.3157943689804271, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4815, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.29591155402427793, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.4704, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.4533644761517681, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.7154, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.27677583362031194, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4838, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.190476190476191, | |
| "grad_norm": 0.27759681038860284, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4567, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.25534553117450187, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.48, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.380952380952381, | |
| "grad_norm": 0.25810299592380054, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4817, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.28200325930565023, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4647, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.2746241939650419, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4775, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.26473488744066087, | |
| "learning_rate": 0.0, | |
| "loss": 0.4753, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "step": 70, | |
| "total_flos": 2.007763554182103e+17, | |
| "train_loss": 0.6434890138251441, | |
| "train_runtime": 3243.4617, | |
| "train_samples_per_second": 2.158, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "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": 2.007763554182103e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |