| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7032967032967034, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014065934065934066, | |
| "grad_norm": 0.6969139575958252, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1559, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014065934065934066, | |
| "eval_loss": 1.1649584770202637, | |
| "eval_runtime": 12.1673, | |
| "eval_samples_per_second": 39.368, | |
| "eval_steps_per_second": 9.862, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.028131868131868132, | |
| "grad_norm": 0.7395194172859192, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1806, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0421978021978022, | |
| "grad_norm": 0.6568499803543091, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 1.1517, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.056263736263736264, | |
| "grad_norm": 0.5321763753890991, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 1.0992, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07032967032967033, | |
| "grad_norm": 0.5334182381629944, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 1.073, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0843956043956044, | |
| "grad_norm": 0.5499136447906494, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 1.0421, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09846153846153846, | |
| "grad_norm": 0.5175103545188904, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 1.0444, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11252747252747253, | |
| "grad_norm": 0.48642808198928833, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.9828, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1265934065934066, | |
| "grad_norm": 0.4232492744922638, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 0.9878, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14065934065934066, | |
| "grad_norm": 0.4139992892742157, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.9824, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15472527472527473, | |
| "grad_norm": 0.3782505691051483, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.9677, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1687912087912088, | |
| "grad_norm": 0.30948084592819214, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.9641, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "grad_norm": 0.2794710099697113, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 0.9449, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "eval_loss": 0.9460345506668091, | |
| "eval_runtime": 11.6744, | |
| "eval_samples_per_second": 41.03, | |
| "eval_steps_per_second": 10.279, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.19692307692307692, | |
| "grad_norm": 0.25389203429222107, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.9444, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.210989010989011, | |
| "grad_norm": 0.2663428485393524, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.9733, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22505494505494505, | |
| "grad_norm": 0.2467697262763977, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.927, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.23912087912087912, | |
| "grad_norm": 0.2407369613647461, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.9409, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2531868131868132, | |
| "grad_norm": 0.2189822942018509, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.9572, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2672527472527473, | |
| "grad_norm": 0.19346760213375092, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.9531, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2813186813186813, | |
| "grad_norm": 0.18553870916366577, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.9292, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2953846153846154, | |
| "grad_norm": 0.19524887204170227, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 0.9341, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.30945054945054945, | |
| "grad_norm": 0.19429989159107208, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.9377, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.32351648351648354, | |
| "grad_norm": 0.19181743264198303, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.9086, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3375824175824176, | |
| "grad_norm": 0.18493668735027313, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.9354, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3516483516483517, | |
| "grad_norm": 0.18037228286266327, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 0.9371, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3657142857142857, | |
| "grad_norm": 0.17428863048553467, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9254, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3657142857142857, | |
| "eval_loss": 0.9181084632873535, | |
| "eval_runtime": 11.6695, | |
| "eval_samples_per_second": 41.047, | |
| "eval_steps_per_second": 10.283, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3797802197802198, | |
| "grad_norm": 0.17674297094345093, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.9264, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.39384615384615385, | |
| "grad_norm": 0.17466501891613007, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.9213, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.40791208791208794, | |
| "grad_norm": 0.16305765509605408, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.9202, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.421978021978022, | |
| "grad_norm": 0.16738246381282806, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.8995, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43604395604395607, | |
| "grad_norm": 0.17170296609401703, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.9352, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4501098901098901, | |
| "grad_norm": 0.1743059605360031, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.9355, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4641758241758242, | |
| "grad_norm": 0.1688753217458725, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.9474, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.47824175824175824, | |
| "grad_norm": 0.1720803827047348, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.9173, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.49230769230769234, | |
| "grad_norm": 0.1710965782403946, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.9272, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5063736263736264, | |
| "grad_norm": 0.16434235870838165, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.9095, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5204395604395604, | |
| "grad_norm": 0.16822023689746857, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.9318, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5345054945054946, | |
| "grad_norm": 0.17256763577461243, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.9447, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5485714285714286, | |
| "grad_norm": 0.17558202147483826, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.9516, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5485714285714286, | |
| "eval_loss": 0.9109680652618408, | |
| "eval_runtime": 11.6756, | |
| "eval_samples_per_second": 41.026, | |
| "eval_steps_per_second": 10.278, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5626373626373626, | |
| "grad_norm": 0.16583934426307678, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.9045, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5767032967032967, | |
| "grad_norm": 0.16519413888454437, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.9248, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5907692307692308, | |
| "grad_norm": 0.16163888573646545, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.8944, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6048351648351649, | |
| "grad_norm": 0.17038829624652863, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 0.9563, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6189010989010989, | |
| "grad_norm": 0.16273872554302216, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.9048, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6329670329670329, | |
| "grad_norm": 0.1728469431400299, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 0.9209, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6470329670329671, | |
| "grad_norm": 0.16880929470062256, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.9191, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.6610989010989011, | |
| "grad_norm": 0.16875362396240234, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.9281, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6751648351648352, | |
| "grad_norm": 0.16660304367542267, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.9525, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6892307692307692, | |
| "grad_norm": 0.16343393921852112, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 0.9133, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7032967032967034, | |
| "grad_norm": 0.1650265008211136, | |
| "learning_rate": 0.0, | |
| "loss": 0.9154, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.7676344066048e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |