| { | |
| "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": 6.712826482876142, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0918, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.739511623001884, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0807, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.278953121448081, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0469, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.8669900448422507, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9509, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.001585981912743, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9338, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.052145985457584, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9517, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 3.861921543938493, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8775, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.0189324103721855, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8848, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.3141205054856906, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8425, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.703775730551971, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8088, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.4788735463276004, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7752, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.4225471178643385, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.758, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.1834286645573922, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7433, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.9689540058068354, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7258, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.1199168770603705, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7317, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.2163483770623886, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7159, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.9583599810476855, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.6991, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 1.183174805535999, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7037, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.8677813420748476, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.6791, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.050027435050424, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.6667, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.8684292087983247, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6566, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 1.0007445298507156, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6176, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.8089549337377572, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6193, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8621946263541884, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6217, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7968563443670871, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6067, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 1.067452354914592, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6046, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7930206763470024, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.5843, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.8899029816550251, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.5989, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.7453958428063189, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.5876, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.6244665578266759, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.5745, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.8257369538618146, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.5931, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5288745486639439, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.5399, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.694884158892613, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5184, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.5910081021708928, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.4974, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.6605065146555008, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5167, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.6452662203875384, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5264, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.6286310862396527, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5275, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.7489679791620766, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.539, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4818005535869053, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.5313, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.6943043210035605, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5197, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.63163957552272, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5182, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.4783404435396881, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5118, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.6422379349603546, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.4531, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.49069284126307844, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.4923, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.4215503007661755, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.4661, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.5178900032033948, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.456, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.5146190973674538, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.4753, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.46237283969211873, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.4876, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.40781558663399636, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.45, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.4340845364187544, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.4635, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.49244600446526865, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.483, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.4191059687523801, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.4675, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.396641986268647, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.4295, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3873763864481581, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4189, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.39907606179617383, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.4403, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.4386003314527719, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4616, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.42627513798813743, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4483, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.3484508218472824, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.4471, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.3436410958659515, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4276, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.3274201503520226, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4526, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.3447652579442819, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.4398, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.3239780015279207, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.447, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.3138785974659222, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4168, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.3059739363994599, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4303, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.29705589718382003, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4119, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.28281294155372466, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4258, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.29864672729335795, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4507, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2905000294782897, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4637, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2969211601116122, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4123, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.27573016273241896, | |
| "learning_rate": 0.0, | |
| "loss": 0.4362, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.1737700574887936e+17, | |
| "train_loss": 0.5947728885071618, | |
| "train_runtime": 8182.4056, | |
| "train_samples_per_second": 0.855, | |
| "train_steps_per_second": 0.009 | |
| } | |
| ], | |
| "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": 2.1737700574887936e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |