| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.848, | |
| "eval_steps": 500, | |
| "global_step": 620, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.8025777339935303, | |
| "learning_rate": 2.2580645161290324e-06, | |
| "loss": 1.4695, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.3798772096633911, | |
| "learning_rate": 5.483870967741935e-06, | |
| "loss": 0.9564, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.1069912910461426, | |
| "learning_rate": 8.70967741935484e-06, | |
| "loss": 0.7467, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.3565958738327026, | |
| "learning_rate": 1.1935483870967743e-05, | |
| "loss": 0.7442, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.8724671006202698, | |
| "learning_rate": 1.5161290322580646e-05, | |
| "loss": 0.6401, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.8387081027030945, | |
| "learning_rate": 1.838709677419355e-05, | |
| "loss": 0.6412, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.112, | |
| "grad_norm": 1.1549522876739502, | |
| "learning_rate": 1.9996038016334953e-05, | |
| "loss": 0.5019, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.272, | |
| "grad_norm": 0.7454654574394226, | |
| "learning_rate": 1.9968952076274873e-05, | |
| "loss": 0.5679, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.432, | |
| "grad_norm": 1.2986727952957153, | |
| "learning_rate": 1.990884868158239e-05, | |
| "loss": 0.5871, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.592, | |
| "grad_norm": 1.4838392734527588, | |
| "learning_rate": 1.9817344551289796e-05, | |
| "loss": 0.5669, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.752, | |
| "grad_norm": 1.1603983640670776, | |
| "learning_rate": 1.9694729658237925e-05, | |
| "loss": 0.5716, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.912, | |
| "grad_norm": 1.4420567750930786, | |
| "learning_rate": 1.954139256400049e-05, | |
| "loss": 0.584, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.064, | |
| "grad_norm": 0.9452749490737915, | |
| "learning_rate": 1.935781918754836e-05, | |
| "loss": 0.4986, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.224, | |
| "grad_norm": 1.114253044128418, | |
| "learning_rate": 1.914459126539224e-05, | |
| "loss": 0.4908, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.384, | |
| "grad_norm": 1.864726185798645, | |
| "learning_rate": 1.8902384508083518e-05, | |
| "loss": 0.4794, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.544, | |
| "grad_norm": 1.430006980895996, | |
| "learning_rate": 1.863196645891518e-05, | |
| "loss": 0.4947, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.7039999999999997, | |
| "grad_norm": 0.9796671271324158, | |
| "learning_rate": 1.8334194061608577e-05, | |
| "loss": 0.4618, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.864, | |
| "grad_norm": 1.1556652784347534, | |
| "learning_rate": 1.8010010944693846e-05, | |
| "loss": 0.4828, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.016, | |
| "grad_norm": 0.9884580373764038, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.4485, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.176, | |
| "grad_norm": 1.4118114709854126, | |
| "learning_rate": 1.7286602283059238e-05, | |
| "loss": 0.4148, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.336, | |
| "grad_norm": 1.833937168121338, | |
| "learning_rate": 1.688966919075687e-05, | |
| "loss": 0.3848, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.496, | |
| "grad_norm": 1.2855585813522339, | |
| "learning_rate": 1.647090301899358e-05, | |
| "loss": 0.3866, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.656, | |
| "grad_norm": 1.1476826667785645, | |
| "learning_rate": 1.60316308206148e-05, | |
| "loss": 0.3989, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.816, | |
| "grad_norm": 1.1333897113800049, | |
| "learning_rate": 1.5573244631224364e-05, | |
| "loss": 0.3955, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.976, | |
| "grad_norm": 1.8493810892105103, | |
| "learning_rate": 1.5097197057880707e-05, | |
| "loss": 0.353, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.128, | |
| "grad_norm": 1.493438720703125, | |
| "learning_rate": 1.4604996675844586e-05, | |
| "loss": 0.3075, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.288, | |
| "grad_norm": 2.338547945022583, | |
| "learning_rate": 1.4098203247965876e-05, | |
| "loss": 0.346, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.448, | |
| "grad_norm": 1.3177692890167236, | |
| "learning_rate": 1.3578422781858994e-05, | |
| "loss": 0.3307, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.608, | |
| "grad_norm": 1.2539737224578857, | |
| "learning_rate": 1.3047302440530538e-05, | |
| "loss": 0.3482, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.768, | |
| "grad_norm": 1.0234341621398926, | |
| "learning_rate": 1.2506525322587207e-05, | |
| "loss": 0.3244, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.928, | |
| "grad_norm": 1.274221658706665, | |
| "learning_rate": 1.1957805128565232e-05, | |
| "loss": 0.278, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 1.353709101676941, | |
| "learning_rate": 1.1402880730283598e-05, | |
| "loss": 0.2497, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "grad_norm": 2.2268123626708984, | |
| "learning_rate": 1.0843510660430447e-05, | |
| "loss": 0.2217, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 2.0857222080230713, | |
| "learning_rate": 1.028146753984505e-05, | |
| "loss": 0.2379, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.5600000000000005, | |
| "grad_norm": 2.352081775665283, | |
| "learning_rate": 9.718532460154948e-06, | |
| "loss": 0.245, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "grad_norm": 1.6790026426315308, | |
| "learning_rate": 9.156489339569555e-06, | |
| "loss": 0.2337, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 0.9887088537216187, | |
| "learning_rate": 8.597119269716403e-06, | |
| "loss": 0.2297, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.032, | |
| "grad_norm": 1.4913004636764526, | |
| "learning_rate": 8.042194871434771e-06, | |
| "loss": 0.2222, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.192, | |
| "grad_norm": 1.9435575008392334, | |
| "learning_rate": 7.493474677412795e-06, | |
| "loss": 0.1506, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.352, | |
| "grad_norm": 1.686815619468689, | |
| "learning_rate": 6.952697559469464e-06, | |
| "loss": 0.1547, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.5120000000000005, | |
| "grad_norm": 1.5592477321624756, | |
| "learning_rate": 6.421577218141007e-06, | |
| "loss": 0.1359, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.672, | |
| "grad_norm": 2.845890760421753, | |
| "learning_rate": 5.901796752034128e-06, | |
| "loss": 0.1817, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.832, | |
| "grad_norm": 1.81553316116333, | |
| "learning_rate": 5.395003324155414e-06, | |
| "loss": 0.135, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.992, | |
| "grad_norm": 1.9674084186553955, | |
| "learning_rate": 4.902802942119293e-06, | |
| "loss": 0.1423, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.144, | |
| "grad_norm": 1.0333631038665771, | |
| "learning_rate": 4.426755368775637e-06, | |
| "loss": 0.0863, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.304, | |
| "grad_norm": 3.6444036960601807, | |
| "learning_rate": 3.9683691793852044e-06, | |
| "loss": 0.0954, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.464, | |
| "grad_norm": 2.4928786754608154, | |
| "learning_rate": 3.5290969810064258e-06, | |
| "loss": 0.0892, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.624, | |
| "grad_norm": 2.4595706462860107, | |
| "learning_rate": 3.110330809243134e-06, | |
| "loss": 0.0705, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.784, | |
| "grad_norm": 2.398029088973999, | |
| "learning_rate": 2.7133977169407634e-06, | |
| "loss": 0.0872, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.944, | |
| "grad_norm": 1.6807390451431274, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.077, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.096, | |
| "grad_norm": 0.9940488934516907, | |
| "learning_rate": 1.9899890553061565e-06, | |
| "loss": 0.0604, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.256, | |
| "grad_norm": 1.9791232347488403, | |
| "learning_rate": 1.6658059383914249e-06, | |
| "loss": 0.0446, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.416, | |
| "grad_norm": 0.7762503027915955, | |
| "learning_rate": 1.3680335410848211e-06, | |
| "loss": 0.0432, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.576, | |
| "grad_norm": 2.0330629348754883, | |
| "learning_rate": 1.097615491916485e-06, | |
| "loss": 0.0428, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.736, | |
| "grad_norm": 0.7456198930740356, | |
| "learning_rate": 8.554087346077633e-07, | |
| "loss": 0.0322, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.896, | |
| "grad_norm": 1.492835283279419, | |
| "learning_rate": 6.421808124516437e-07, | |
| "loss": 0.0402, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 9.048, | |
| "grad_norm": 0.904043972492218, | |
| "learning_rate": 4.5860743599951186e-07, | |
| "loss": 0.0288, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 9.208, | |
| "grad_norm": 0.47524094581604004, | |
| "learning_rate": 3.0527034176207724e-07, | |
| "loss": 0.029, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.368, | |
| "grad_norm": 1.197879433631897, | |
| "learning_rate": 1.8265544871020724e-07, | |
| "loss": 0.0222, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.528, | |
| "grad_norm": 0.608761727809906, | |
| "learning_rate": 9.11513184176116e-08, | |
| "loss": 0.0235, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.688, | |
| "grad_norm": 0.5942819714546204, | |
| "learning_rate": 3.104792372512822e-08, | |
| "loss": 0.0249, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.848, | |
| "grad_norm": 0.9274442791938782, | |
| "learning_rate": 2.535729828669897e-09, | |
| "loss": 0.0308, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.848, | |
| "step": 620, | |
| "total_flos": 4.910594218000384e+16, | |
| "train_loss": 0.3172732482754415, | |
| "train_runtime": 12792.4734, | |
| "train_samples_per_second": 0.391, | |
| "train_steps_per_second": 0.048 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 620, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.910594218000384e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |