| { | |
| "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": 6.499798718528108, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.0502, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 6.358074852494718, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0201, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 5.901850143089409, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 1.0202, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 2.828555620770347, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.9555, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.9917973015401658, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.9447, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 4.089296383633065, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.9383, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 4.050084231740121, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8904, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.8685075980113237, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.8775, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 2.16321192187975, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.8313, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.6577158839963577, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.8059, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 1.2924183315507582, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.7918, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 1.1540206032085756, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.7539, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 1.219961272326993, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.7491, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 1.1253688956814922, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7412, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.009233707389781, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.7346, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 1.0873337183437726, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.7268, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 1.0721507507564518, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.7157, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.9648884964828792, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.7233, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.9668012402464797, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.7066, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.934154371642649, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.7161, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.7166389634123703, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.6909, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.8710853410931203, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.6475, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.7503997657395542, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.6636, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.8477977423578985, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.6467, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.7186258276657341, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.6541, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.8058538702291377, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.6333, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.589177415182319, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.6312, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6270720796663206, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.6299, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5779943489596228, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.6349, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5801587235924598, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.6334, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.5191037047749684, | |
| "learning_rate": 1.3653410243663953e-05, | |
| "loss": 0.6171, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.6215837624599388, | |
| "learning_rate": 1.3184866502516846e-05, | |
| "loss": 0.6123, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.48035200514141696, | |
| "learning_rate": 1.2708404681430054e-05, | |
| "loss": 0.5799, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.2640000000000002, | |
| "grad_norm": 0.4837424313330577, | |
| "learning_rate": 1.2225209339563144e-05, | |
| "loss": 0.5995, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.499636973382339, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.5771, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.456, | |
| "grad_norm": 0.43396701626690587, | |
| "learning_rate": 1.1243437046474854e-05, | |
| "loss": 0.5574, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.552, | |
| "grad_norm": 0.5013651998235505, | |
| "learning_rate": 1.0747300935864245e-05, | |
| "loss": 0.5787, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.648, | |
| "grad_norm": 0.429973260279144, | |
| "learning_rate": 1.0249306917380731e-05, | |
| "loss": 0.5707, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.7439999999999998, | |
| "grad_norm": 0.43334786568274714, | |
| "learning_rate": 9.750693082619274e-06, | |
| "loss": 0.567, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.4384329686259764, | |
| "learning_rate": 9.252699064135759e-06, | |
| "loss": 0.5646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.936, | |
| "grad_norm": 0.4620434529258523, | |
| "learning_rate": 8.756562953525151e-06, | |
| "loss": 0.573, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.032, | |
| "grad_norm": 0.43889335863783374, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.541, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 0.38711335201171704, | |
| "learning_rate": 7.774790660436857e-06, | |
| "loss": 0.5353, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.224, | |
| "grad_norm": 0.41300190240021895, | |
| "learning_rate": 7.291595318569951e-06, | |
| "loss": 0.5386, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.38554343062473856, | |
| "learning_rate": 6.815133497483157e-06, | |
| "loss": 0.5356, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.416, | |
| "grad_norm": 0.3774706316273223, | |
| "learning_rate": 6.34658975633605e-06, | |
| "loss": 0.541, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.5120000000000005, | |
| "grad_norm": 0.34950120520525385, | |
| "learning_rate": 5.887128968693887e-06, | |
| "loss": 0.5262, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 0.4038357155214054, | |
| "learning_rate": 5.43789342646837e-06, | |
| "loss": 0.5386, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.704, | |
| "grad_norm": 0.3354570207793457, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.5355, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3036598935840911, | |
| "learning_rate": 4.5745373613424075e-06, | |
| "loss": 0.5127, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.896, | |
| "grad_norm": 0.3191245238152673, | |
| "learning_rate": 4.162563277652104e-06, | |
| "loss": 0.5289, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 4.992, | |
| "grad_norm": 0.3149688192700122, | |
| "learning_rate": 3.7651019814126656e-06, | |
| "loss": 0.5128, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.088, | |
| "grad_norm": 0.3163261190560095, | |
| "learning_rate": 3.3831416240314085e-06, | |
| "loss": 0.5114, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.184, | |
| "grad_norm": 0.30956344837523775, | |
| "learning_rate": 3.017631819139273e-06, | |
| "loss": 0.5088, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.27290630038585995, | |
| "learning_rate": 2.669481281701739e-06, | |
| "loss": 0.5087, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.376, | |
| "grad_norm": 0.260625050034014, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.5034, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.4719999999999995, | |
| "grad_norm": 0.28556263030291057, | |
| "learning_rate": 2.0286749277707783e-06, | |
| "loss": 0.503, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 5.568, | |
| "grad_norm": 0.2726932620433162, | |
| "learning_rate": 1.7376122568400533e-06, | |
| "loss": 0.5078, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.664, | |
| "grad_norm": 0.2635992369702704, | |
| "learning_rate": 1.467091183678444e-06, | |
| "loss": 0.4944, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.2900005697643862, | |
| "learning_rate": 1.2177842662977136e-06, | |
| "loss": 0.4998, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.856, | |
| "grad_norm": 0.2785091103108078, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.502, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 5.952, | |
| "grad_norm": 0.26356221649060024, | |
| "learning_rate": 7.852378812959227e-07, | |
| "loss": 0.5094, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.048, | |
| "grad_norm": 0.26595934833727786, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.4961, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.144, | |
| "grad_norm": 0.2478902264268863, | |
| "learning_rate": 4.4427194213859216e-07, | |
| "loss": 0.5088, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "grad_norm": 0.25456009362039866, | |
| "learning_rate": 3.0922713770922155e-07, | |
| "loss": 0.4883, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 6.336, | |
| "grad_norm": 0.23469717172618107, | |
| "learning_rate": 1.9827512151456175e-07, | |
| "loss": 0.5004, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.432, | |
| "grad_norm": 0.2480253059896632, | |
| "learning_rate": 1.1169173774871478e-07, | |
| "loss": 0.4909, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 6.5280000000000005, | |
| "grad_norm": 0.24442728283516682, | |
| "learning_rate": 4.9692246345985905e-08, | |
| "loss": 0.5014, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.624, | |
| "grad_norm": 0.2513696621604786, | |
| "learning_rate": 1.2430787810776556e-08, | |
| "loss": 0.4806, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 0.23965528855214976, | |
| "learning_rate": 0.0, | |
| "loss": 0.5013, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 70, | |
| "total_flos": 2.8997918553381274e+17, | |
| "train_loss": 0.6345526273761477, | |
| "train_runtime": 9190.0056, | |
| "train_samples_per_second": 0.762, | |
| "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": 2.8997918553381274e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |