| { | |
| "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": 6.449225177837263, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0322, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.602005389436255, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0641, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.835537225946153, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0215, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.8022376676677316, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.955, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.054369441478321, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9322, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.118727746013796, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.944, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.072757437093746, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9158, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.9086999603728185, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8825, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.0457921871294236, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8555, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.58641699779283, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8191, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.113820530474924, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7885, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1993205495842958, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7629, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.2676894042742657, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7689, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.2170547392385078, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7608, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.143208755030752, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7532, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.9489471688862002, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7537, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.938789242293279, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7189, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.8444639518721001, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7183, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.7904244208592369, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7113, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6234032641195104, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7085, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.7262675713838688, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6952, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.6120496173094568, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6716, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7096145713057932, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6555, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.5770319140453306, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.658, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6580001511060984, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.6774651525355592, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6488, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5237012374177746, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6397, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.5989082356134461, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6477, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.4749997317975513, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6229, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5794522762200444, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.534929837544779, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6205, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.4307173134390121, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6075, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.5511884845985844, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5914, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.38888174563748434, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5732, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.41269985584062174, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5703, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.5516548859697321, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5851, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.46043552563191753, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5962, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.46239281444386476, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5782, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.4683150754553442, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.574, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.48315307736080054, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5601, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.458649949913566, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.5673, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.412728039938624, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.5655, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.5095825216875302, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5387, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.44123669074359173, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5444, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.39254009120718975, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5435, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.40347401177250125, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.5304, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.36214698559093583, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5202, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.3753554059852131, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5011, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.37737194348256425, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.547, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3379625911432947, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5212, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.33700973019549957, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5275, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.34957313687729963, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5189, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3731827927907244, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5069, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.31666320804581394, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.4871, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.29438736514197705, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.5072, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.2933292929722914, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.4986, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.29992331032812136, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.503, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2769240103107844, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5223, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.2854750359849159, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4962, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.28968762442683227, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.502, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.2889062667721945, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.5016, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.2667628399615331, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5109, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.2575360045289637, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.5078, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2448136152113595, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.4999, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.29676714676627197, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.5037, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.24716299784402054, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.4851, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.23877888538515246, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.5093, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.25403530162126076, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.4776, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2509774030133356, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4986, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.23917547182725768, | |
| "learning_rate": 0.0, | |
| "loss": 0.4928, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.8410057313170227e+17, | |
| "train_loss": 0.638406787599836, | |
| "train_runtime": 8875.7256, | |
| "train_samples_per_second": 0.789, | |
| "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": 2.8410057313170227e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |