| { | |
| "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": 6.482630047676832, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.048, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.605927045389436, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0799, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 6.0004200415773505, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0442, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.734011590523821, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9666, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.84249422805747, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9537, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.332528721513862, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.963, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.30730089643909, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9463, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 3.0689171204658607, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.9031, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.1787532897192405, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.851, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.5726659771222042, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8165, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.245544037171024, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.8097, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.3131832085531985, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7686, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.3383824192343372, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7878, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.1553682192981205, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7525, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.0896390319471496, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7737, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.3337373183843793, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7428, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.013522026733598, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7436, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9815484856530986, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7362, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7790354467718542, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7393, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.7658882104136794, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.738, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.7947036239545487, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.718, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.852766632181634, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6934, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7394983824426842, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.666, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.7639113743189007, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6736, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6699889236649792, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6612, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.7664583521668237, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6439, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.7078267408855885, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6707, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5967666407530697, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6389, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.6501291337964047, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6453, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5499670913883455, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6531, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.6575754490622177, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6383, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.5555064813802514, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6244, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5496645892387174, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5952, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4637803134181456, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5929, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.5112699616498038, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5952, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.43277176797556094, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5883, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5222777088315609, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5908, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.41973649970207644, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5882, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4342864360519182, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.6026, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.381967825998277, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5629, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.44818691951993556, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5778, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.3885131368709949, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5831, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.4447170619083982, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.562, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.37010318797437036, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5516, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.3470059178314434, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5467, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.3988709710819742, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5396, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.4221839341202327, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5606, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.3625209658466445, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5385, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.36454061846465446, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5348, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.373324742871357, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5275, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3573424788450633, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5297, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.36411889766130145, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5286, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3734920149644589, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5133, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.3412001600297152, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5186, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.31407697539213053, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.5425, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.3114038099525229, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5184, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.30769049378824825, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.4957, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.30020372385344285, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5117, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.28151111678907115, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.5186, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.286664292344187, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.504, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.28166817278860223, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5127, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2763270403009709, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5331, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.2841023449050398, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4907, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2847352132499023, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4897, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.23751825841967308, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.5173, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.25459515583756376, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.5033, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.23460332894810443, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.5221, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.2445021863007875, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.5106, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.23921124754838127, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.5092, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.2517352572798992, | |
| "learning_rate": 0.0, | |
| "loss": 0.5133, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.8492269037079757e+17, | |
| "train_loss": 0.6516103263412203, | |
| "train_runtime": 9016.1191, | |
| "train_samples_per_second": 0.776, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "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": 2.8492269037079757e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |