| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 70, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.991291106423771, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0604, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.9404392087473, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0566, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.417751052511398, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0483, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.242138270087566, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9819, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.267046592642321, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9495, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.449643412615094, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9476, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.2543563479887805, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8945, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.04655925911499, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8817, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.6061134673928112, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8523, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8345700681584693, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8179, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.6553945729979291, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.79, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.7435599135273063, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7653, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.285420005987591, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7539, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.4808992380192267, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7555, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.2689474653555437, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7299, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.3090980059946602, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7253, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.0613800408829623, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7176, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.0428487681920615, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7211, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 1.009791001357855, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6964, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8816529724390634, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7113, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.9551087649605005, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.7026, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.7627964904216537, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6638, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.848678521181871, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6567, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.6539475241251977, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6487, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.8990612791765852, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6344, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7416213329212227, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6484, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7328908465586557, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6299, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.7206262580746776, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6169, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6434789028128752, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6405, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6347294733542639, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6093, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.6152630596334819, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6015, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.748786730756543, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5864, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.48685398125035, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5867, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5373564725675727, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5782, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.535644151595081, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5728, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.44261470044848933, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5304, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.48019507781488413, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5532, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.43770022949236825, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5714, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4910501759018241, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5547, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.48029309503660694, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5452, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.46085280409568424, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5773, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.4369558520317486, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5598, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5072993680368153, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5164, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.3773968253479734, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5408, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4481137871258331, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.4941, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.47018406908380456, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5284, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5171527881680771, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5014, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.39260580816991114, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.4112236644270957, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.4793, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.40770767189599294, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5223, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.4302454987757317, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5474, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.3557016706548058, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5191, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.4114725591225338, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4818, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3754167163800502, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.474, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.32245438031841833, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4917, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.34296759997920345, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4908, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.3902739816060275, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4724, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.35043080542410515, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4812, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.30341222784616684, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4915, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.32931258071115643, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4941, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.334337445206413, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4871, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3088070421813325, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5032, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3594182587859646, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4605, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.3603587818182848, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4592, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.28400234193835516, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4724, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.27318126327227227, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4849, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.35424266564159845, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4629, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2619950940084347, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4865, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.25288925538925616, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4741, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.27245162033242015, | |
| "learning_rate": 0.0, | |
| "loss": 0.4932, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.46611802945749e+17, | |
| "train_loss": 0.6276680737733841, | |
| "train_runtime": 7354.819, | |
| "train_samples_per_second": 0.952, | |
| "train_steps_per_second": 0.01 | |
| } | |
| ], | |
| "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": 2.46611802945749e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |