| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 52.583274841308594, | |
| "learning_rate": 1.57035175879397e-09, | |
| "loss": 4.4832, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 52.1717414855957, | |
| "learning_rate": 3.14070351758794e-09, | |
| "loss": 4.4526, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 51.840389251708984, | |
| "learning_rate": 4.71105527638191e-09, | |
| "loss": 4.4294, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 52.11858367919922, | |
| "learning_rate": 6.28140703517588e-09, | |
| "loss": 4.4392, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 51.991600036621094, | |
| "learning_rate": 7.85175879396985e-09, | |
| "loss": 4.4464, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 52.67415237426758, | |
| "learning_rate": 9.42211055276382e-09, | |
| "loss": 4.4485, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 52.06214904785156, | |
| "learning_rate": 1.099246231155779e-08, | |
| "loss": 4.445, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 52.24057388305664, | |
| "learning_rate": 1.256281407035176e-08, | |
| "loss": 4.4631, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 4.591891765594482, | |
| "eval_runtime": 48.4152, | |
| "eval_samples_per_second": 28.214, | |
| "eval_steps_per_second": 0.289, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 52.291419982910156, | |
| "learning_rate": 1.413316582914573e-08, | |
| "loss": 4.4435, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 51.911685943603516, | |
| "learning_rate": 1.57035175879397e-08, | |
| "loss": 4.4448, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 52.54676055908203, | |
| "learning_rate": 1.7273869346733672e-08, | |
| "loss": 4.4468, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 52.45462417602539, | |
| "learning_rate": 1.884422110552764e-08, | |
| "loss": 4.4229, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 52.290565490722656, | |
| "learning_rate": 2.041457286432161e-08, | |
| "loss": 4.4212, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 51.737483978271484, | |
| "learning_rate": 2.198492462311558e-08, | |
| "loss": 4.4379, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 52.21713638305664, | |
| "learning_rate": 2.3555276381909547e-08, | |
| "loss": 4.426, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 52.21146011352539, | |
| "learning_rate": 2.512562814070352e-08, | |
| "loss": 4.4206, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 4.538872241973877, | |
| "eval_runtime": 45.8583, | |
| "eval_samples_per_second": 29.787, | |
| "eval_steps_per_second": 0.305, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 52.31325149536133, | |
| "learning_rate": 2.669597989949749e-08, | |
| "loss": 4.4286, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 52.39244842529297, | |
| "learning_rate": 2.826633165829146e-08, | |
| "loss": 4.4363, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.375, | |
| "grad_norm": 51.97914123535156, | |
| "learning_rate": 2.983668341708543e-08, | |
| "loss": 4.4068, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 52.424015045166016, | |
| "learning_rate": 3.14070351758794e-08, | |
| "loss": 4.4079, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.625, | |
| "grad_norm": 51.720726013183594, | |
| "learning_rate": 3.297738693467337e-08, | |
| "loss": 4.4012, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 51.92622375488281, | |
| "learning_rate": 3.4547738693467345e-08, | |
| "loss": 4.3656, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 51.43586730957031, | |
| "learning_rate": 3.611809045226131e-08, | |
| "loss": 4.3986, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 52.25431823730469, | |
| "learning_rate": 3.768844221105528e-08, | |
| "loss": 4.3742, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 4.466385841369629, | |
| "eval_runtime": 43.0073, | |
| "eval_samples_per_second": 31.762, | |
| "eval_steps_per_second": 0.326, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 52.38350296020508, | |
| "learning_rate": 3.925879396984925e-08, | |
| "loss": 4.3569, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 52.09661865234375, | |
| "learning_rate": 4.082914572864322e-08, | |
| "loss": 4.3329, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.375, | |
| "grad_norm": 52.156551361083984, | |
| "learning_rate": 4.239949748743719e-08, | |
| "loss": 4.3611, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 51.66514587402344, | |
| "learning_rate": 4.396984924623116e-08, | |
| "loss": 4.3414, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.625, | |
| "grad_norm": 51.9528694152832, | |
| "learning_rate": 4.5540201005025126e-08, | |
| "loss": 4.3208, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 51.63153839111328, | |
| "learning_rate": 4.7110552763819094e-08, | |
| "loss": 4.3323, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.875, | |
| "grad_norm": 51.82636642456055, | |
| "learning_rate": 4.8680904522613075e-08, | |
| "loss": 4.2969, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 50.99885177612305, | |
| "learning_rate": 5.025125628140704e-08, | |
| "loss": 4.2848, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 4.35823392868042, | |
| "eval_runtime": 42.4893, | |
| "eval_samples_per_second": 32.149, | |
| "eval_steps_per_second": 0.329, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.125, | |
| "grad_norm": 51.07362365722656, | |
| "learning_rate": 5.182160804020101e-08, | |
| "loss": 4.2605, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 51.94071578979492, | |
| "learning_rate": 5.339195979899498e-08, | |
| "loss": 4.2979, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 50.928531646728516, | |
| "learning_rate": 5.496231155778895e-08, | |
| "loss": 4.2821, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 51.46839141845703, | |
| "learning_rate": 5.653266331658292e-08, | |
| "loss": 4.2471, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.625, | |
| "grad_norm": 51.3604621887207, | |
| "learning_rate": 5.810301507537689e-08, | |
| "loss": 4.2608, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 51.4367561340332, | |
| "learning_rate": 5.967336683417086e-08, | |
| "loss": 4.24, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.875, | |
| "grad_norm": 51.37506103515625, | |
| "learning_rate": 6.124371859296483e-08, | |
| "loss": 4.2268, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 51.0534782409668, | |
| "learning_rate": 6.28140703517588e-08, | |
| "loss": 4.1992, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 4.225412845611572, | |
| "eval_runtime": 62.0906, | |
| "eval_samples_per_second": 22.0, | |
| "eval_steps_per_second": 0.225, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.125, | |
| "grad_norm": 50.854610443115234, | |
| "learning_rate": 6.438442211055277e-08, | |
| "loss": 4.1707, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 51.0251579284668, | |
| "learning_rate": 6.595477386934674e-08, | |
| "loss": 4.1889, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.375, | |
| "grad_norm": 51.27962875366211, | |
| "learning_rate": 6.75251256281407e-08, | |
| "loss": 4.1892, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 50.632904052734375, | |
| "learning_rate": 6.909547738693469e-08, | |
| "loss": 4.1684, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 51.070587158203125, | |
| "learning_rate": 7.066582914572865e-08, | |
| "loss": 4.154, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 50.42203903198242, | |
| "learning_rate": 7.223618090452263e-08, | |
| "loss": 4.1407, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.875, | |
| "grad_norm": 50.98355484008789, | |
| "learning_rate": 7.380653266331659e-08, | |
| "loss": 4.129, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 50.823097229003906, | |
| "learning_rate": 7.537688442211056e-08, | |
| "loss": 4.1172, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 4.049016952514648, | |
| "eval_runtime": 46.6978, | |
| "eval_samples_per_second": 29.252, | |
| "eval_steps_per_second": 0.3, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.125, | |
| "grad_norm": 51.02920150756836, | |
| "learning_rate": 7.694723618090454e-08, | |
| "loss": 4.1099, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 50.04145812988281, | |
| "learning_rate": 7.85175879396985e-08, | |
| "loss": 4.076, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.375, | |
| "grad_norm": 50.59613037109375, | |
| "learning_rate": 8.008793969849247e-08, | |
| "loss": 4.0718, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 50.145957946777344, | |
| "learning_rate": 8.165829145728645e-08, | |
| "loss": 4.0557, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.625, | |
| "grad_norm": 49.8894157409668, | |
| "learning_rate": 8.322864321608041e-08, | |
| "loss": 4.0178, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 49.86162185668945, | |
| "learning_rate": 8.479899497487438e-08, | |
| "loss": 4.0299, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 50.02029800415039, | |
| "learning_rate": 8.636934673366834e-08, | |
| "loss": 3.9924, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 49.66703414916992, | |
| "learning_rate": 8.793969849246232e-08, | |
| "loss": 3.9685, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 3.8612220287323, | |
| "eval_runtime": 46.1098, | |
| "eval_samples_per_second": 29.625, | |
| "eval_steps_per_second": 0.304, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.125, | |
| "grad_norm": 49.38985824584961, | |
| "learning_rate": 8.951005025125629e-08, | |
| "loss": 3.9575, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 49.38282012939453, | |
| "learning_rate": 9.108040201005025e-08, | |
| "loss": 3.9457, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 7.375, | |
| "grad_norm": 49.578147888183594, | |
| "learning_rate": 9.265075376884423e-08, | |
| "loss": 3.9147, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 49.263980865478516, | |
| "learning_rate": 9.422110552763819e-08, | |
| "loss": 3.9124, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.625, | |
| "grad_norm": 48.740360260009766, | |
| "learning_rate": 9.579145728643216e-08, | |
| "loss": 3.9056, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 49.30016326904297, | |
| "learning_rate": 9.736180904522615e-08, | |
| "loss": 3.8962, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.875, | |
| "grad_norm": 48.78429412841797, | |
| "learning_rate": 9.89321608040201e-08, | |
| "loss": 3.8614, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 48.219146728515625, | |
| "learning_rate": 1.0050251256281409e-07, | |
| "loss": 3.8165, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.6316096782684326, | |
| "eval_runtime": 45.5729, | |
| "eval_samples_per_second": 29.974, | |
| "eval_steps_per_second": 0.307, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 48.37006759643555, | |
| "learning_rate": 1.0207286432160806e-07, | |
| "loss": 3.8137, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 47.9218635559082, | |
| "learning_rate": 1.0364321608040202e-07, | |
| "loss": 3.7862, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.375, | |
| "grad_norm": 48.484378814697266, | |
| "learning_rate": 1.05213567839196e-07, | |
| "loss": 3.789, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 48.121212005615234, | |
| "learning_rate": 1.0678391959798996e-07, | |
| "loss": 3.7609, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.625, | |
| "grad_norm": 48.182281494140625, | |
| "learning_rate": 1.0835427135678393e-07, | |
| "loss": 3.7594, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 48.09780502319336, | |
| "learning_rate": 1.099246231155779e-07, | |
| "loss": 3.6978, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.875, | |
| "grad_norm": 47.829345703125, | |
| "learning_rate": 1.1149497487437187e-07, | |
| "loss": 3.7101, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 47.34735107421875, | |
| "learning_rate": 1.1306532663316584e-07, | |
| "loss": 3.6672, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.3924500942230225, | |
| "eval_runtime": 42.4081, | |
| "eval_samples_per_second": 32.211, | |
| "eval_steps_per_second": 0.33, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.125, | |
| "grad_norm": 47.713768005371094, | |
| "learning_rate": 1.146356783919598e-07, | |
| "loss": 3.6657, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 48.04665756225586, | |
| "learning_rate": 1.1620603015075378e-07, | |
| "loss": 3.635, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 47.147972106933594, | |
| "learning_rate": 1.1777638190954775e-07, | |
| "loss": 3.6149, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 47.14889144897461, | |
| "learning_rate": 1.193467336683417e-07, | |
| "loss": 3.5891, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.625, | |
| "grad_norm": 47.27253723144531, | |
| "learning_rate": 1.209170854271357e-07, | |
| "loss": 3.56, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 47.30030822753906, | |
| "learning_rate": 1.2248743718592966e-07, | |
| "loss": 3.554, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.875, | |
| "grad_norm": 46.53535461425781, | |
| "learning_rate": 1.2405778894472362e-07, | |
| "loss": 3.5081, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 46.524356842041016, | |
| "learning_rate": 1.256281407035176e-07, | |
| "loss": 3.4906, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.130697011947632, | |
| "eval_runtime": 44.7299, | |
| "eval_samples_per_second": 30.539, | |
| "eval_steps_per_second": 0.313, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 0.0, | |
| "train_batch_size": 100, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |