| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03221908981071285, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006443817962142569, | |
| "grad_norm": 4.898526668548584, | |
| "learning_rate": 1e-05, | |
| "loss": 8.2457, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006443817962142569, | |
| "eval_loss": 7.802831649780273, | |
| "eval_runtime": 47.5305, | |
| "eval_samples_per_second": 54.996, | |
| "eval_steps_per_second": 27.498, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012887635924285139, | |
| "grad_norm": 3.692995548248291, | |
| "learning_rate": 2e-05, | |
| "loss": 8.3769, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0019331453886427708, | |
| "grad_norm": 4.796352386474609, | |
| "learning_rate": 3e-05, | |
| "loss": 8.6187, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0025775271848570277, | |
| "grad_norm": 3.221654176712036, | |
| "learning_rate": 4e-05, | |
| "loss": 8.0322, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003221908981071285, | |
| "grad_norm": 2.433626174926758, | |
| "learning_rate": 5e-05, | |
| "loss": 7.9877, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0038662907772855416, | |
| "grad_norm": 2.0175940990448, | |
| "learning_rate": 6e-05, | |
| "loss": 7.9417, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004510672573499798, | |
| "grad_norm": 2.2974374294281006, | |
| "learning_rate": 7e-05, | |
| "loss": 8.3393, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0051550543697140554, | |
| "grad_norm": 1.9051103591918945, | |
| "learning_rate": 8e-05, | |
| "loss": 7.4193, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005799436165928313, | |
| "grad_norm": 2.320819616317749, | |
| "learning_rate": 9e-05, | |
| "loss": 8.0907, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00644381796214257, | |
| "grad_norm": 1.9515935182571411, | |
| "learning_rate": 0.0001, | |
| "loss": 7.6772, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007088199758356826, | |
| "grad_norm": 1.723854899406433, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 7.4406, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007732581554571083, | |
| "grad_norm": 3.2257795333862305, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 7.8407, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008376963350785341, | |
| "grad_norm": 2.5419061183929443, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 8.0091, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008376963350785341, | |
| "eval_loss": 7.61268949508667, | |
| "eval_runtime": 23.5679, | |
| "eval_samples_per_second": 110.914, | |
| "eval_steps_per_second": 55.457, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009021345146999597, | |
| "grad_norm": 3.3199102878570557, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 7.7896, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009665726943213854, | |
| "grad_norm": 3.909156322479248, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 7.9258, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010310108739428111, | |
| "grad_norm": 2.004795551300049, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 7.6277, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.010954490535642368, | |
| "grad_norm": 1.8555487394332886, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 7.9832, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011598872331856625, | |
| "grad_norm": 2.1525354385375977, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 7.4556, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012243254128070882, | |
| "grad_norm": 2.464871406555176, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 7.3449, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.01288763592428514, | |
| "grad_norm": 1.7199599742889404, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.4561, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013532017720499397, | |
| "grad_norm": 1.4880601167678833, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 7.7051, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014176399516713652, | |
| "grad_norm": 2.7781991958618164, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 7.8179, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01482078131292791, | |
| "grad_norm": 1.7467265129089355, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 7.6382, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.015465163109142166, | |
| "grad_norm": 2.2026655673980713, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 7.5733, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.016109544905356425, | |
| "grad_norm": 1.6499930620193481, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 8.1215, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.016753926701570682, | |
| "grad_norm": 3.187775135040283, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 7.2734, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.016753926701570682, | |
| "eval_loss": 7.397295951843262, | |
| "eval_runtime": 23.7029, | |
| "eval_samples_per_second": 110.282, | |
| "eval_steps_per_second": 55.141, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.017398308497784936, | |
| "grad_norm": 1.248848557472229, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 7.0013, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.018042690293999193, | |
| "grad_norm": 1.4550312757492065, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 7.6238, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01868707209021345, | |
| "grad_norm": 2.163661241531372, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 8.13, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.019331453886427707, | |
| "grad_norm": 1.2935118675231934, | |
| "learning_rate": 5e-05, | |
| "loss": 7.4064, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.019975835682641965, | |
| "grad_norm": 1.2071995735168457, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 7.492, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.020620217478856222, | |
| "grad_norm": 1.0634459257125854, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 7.3088, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02126459927507048, | |
| "grad_norm": 1.2333537340164185, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 7.4175, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.021908981071284736, | |
| "grad_norm": 1.3628766536712646, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 7.3945, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.022553362867498993, | |
| "grad_norm": 1.8400943279266357, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 7.7443, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02319774466371325, | |
| "grad_norm": 1.8728278875350952, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 7.7232, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.023842126459927507, | |
| "grad_norm": 1.2013438940048218, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 7.6996, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.024486508256141765, | |
| "grad_norm": 2.287853240966797, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 7.2294, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.025130890052356022, | |
| "grad_norm": 1.4032453298568726, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 7.8582, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.025130890052356022, | |
| "eval_loss": 7.3332672119140625, | |
| "eval_runtime": 23.6689, | |
| "eval_samples_per_second": 110.44, | |
| "eval_steps_per_second": 55.22, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.02577527184857028, | |
| "grad_norm": 2.051464796066284, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 7.3642, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.026419653644784536, | |
| "grad_norm": 1.3412200212478638, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 7.4789, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.027064035440998793, | |
| "grad_norm": 1.0841667652130127, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 7.6536, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.02770841723721305, | |
| "grad_norm": 1.1670408248901367, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 7.7881, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.028352799033427304, | |
| "grad_norm": 2.0704867839813232, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 7.6421, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02899718082964156, | |
| "grad_norm": 1.0424221754074097, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 7.3945, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02964156262585582, | |
| "grad_norm": 1.2187137603759766, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 7.8543, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.030285944422070075, | |
| "grad_norm": 1.103108286857605, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 7.4402, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.030930326218284333, | |
| "grad_norm": 1.0252392292022705, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 7.3845, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03157470801449859, | |
| "grad_norm": 1.0439541339874268, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 7.4411, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03221908981071285, | |
| "grad_norm": 1.1140565872192383, | |
| "learning_rate": 0.0, | |
| "loss": 7.2666, | |
| "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": 109141070708736.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |