| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00330515687100799, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.61031374201598e-05, | |
| "grad_norm": 4.053699016571045, | |
| "learning_rate": 1e-05, | |
| "loss": 31.7526, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.61031374201598e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1234.9622, | |
| "eval_samples_per_second": 20.631, | |
| "eval_steps_per_second": 10.316, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001322062748403196, | |
| "grad_norm": 3.7032253742218018, | |
| "learning_rate": 2e-05, | |
| "loss": 28.2513, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00019830941226047942, | |
| "grad_norm": 3.8769631385803223, | |
| "learning_rate": 3e-05, | |
| "loss": 28.1183, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002644125496806392, | |
| "grad_norm": 3.5392589569091797, | |
| "learning_rate": 4e-05, | |
| "loss": 23.6722, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00033051568710079903, | |
| "grad_norm": 3.854933023452759, | |
| "learning_rate": 5e-05, | |
| "loss": 23.7722, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00039661882452095884, | |
| "grad_norm": 3.8566882610321045, | |
| "learning_rate": 6e-05, | |
| "loss": 22.6891, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00046272196194111864, | |
| "grad_norm": 5.320549964904785, | |
| "learning_rate": 7e-05, | |
| "loss": 27.2562, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0005288250993612784, | |
| "grad_norm": 5.6749186515808105, | |
| "learning_rate": 8e-05, | |
| "loss": 26.0755, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0005949282367814382, | |
| "grad_norm": 7.9273223876953125, | |
| "learning_rate": 9e-05, | |
| "loss": 23.9769, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006610313742015981, | |
| "grad_norm": 7.923571586608887, | |
| "learning_rate": 0.0001, | |
| "loss": 24.9361, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007271345116217578, | |
| "grad_norm": 9.730711936950684, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 23.4156, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0007932376490419177, | |
| "grad_norm": 10.039743423461914, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 24.2207, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008593407864620774, | |
| "grad_norm": 9.729897499084473, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 21.1039, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0008593407864620774, | |
| "eval_loss": NaN, | |
| "eval_runtime": 972.9875, | |
| "eval_samples_per_second": 26.186, | |
| "eval_steps_per_second": 13.094, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0009254439238822373, | |
| "grad_norm": 10.284782409667969, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 19.3567, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0009915470613023971, | |
| "grad_norm": 7.863360404968262, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 15.71, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001057650198722557, | |
| "grad_norm": 5.936141490936279, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 18.2659, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0011237533361427166, | |
| "grad_norm": 13.269340515136719, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 15.3991, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0011898564735628764, | |
| "grad_norm": 10.018537521362305, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 16.9379, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012559596109830364, | |
| "grad_norm": 10.708539009094238, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 18.7077, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0013220627484031961, | |
| "grad_norm": 8.985401153564453, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 17.9045, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0013881658858233559, | |
| "grad_norm": 7.074049949645996, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 16.0002, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0014542690232435156, | |
| "grad_norm": 5.58726167678833, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 14.2095, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0015203721606636756, | |
| "grad_norm": 6.3829874992370605, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 15.7749, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0015864752980838353, | |
| "grad_norm": 6.384377479553223, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 17.0108, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001652578435503995, | |
| "grad_norm": 6.838883399963379, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 19.6889, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0017186815729241548, | |
| "grad_norm": 6.230001926422119, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 17.9956, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0017186815729241548, | |
| "eval_loss": NaN, | |
| "eval_runtime": 973.9081, | |
| "eval_samples_per_second": 26.162, | |
| "eval_steps_per_second": 13.081, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0017847847103443148, | |
| "grad_norm": 6.071116924285889, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 18.2956, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0018508878477644746, | |
| "grad_norm": 7.337187767028809, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 17.3422, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0019169909851846343, | |
| "grad_norm": 7.344765663146973, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 17.1084, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0019830941226047943, | |
| "grad_norm": 5.357908248901367, | |
| "learning_rate": 5e-05, | |
| "loss": 16.2214, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002049197260024954, | |
| "grad_norm": 5.5231170654296875, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 14.8315, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.002115300397445114, | |
| "grad_norm": 6.47795295715332, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 18.0969, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0021814035348652733, | |
| "grad_norm": 7.260648250579834, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 15.8214, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0022475066722854333, | |
| "grad_norm": 7.220073699951172, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 12.7376, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0023136098097055933, | |
| "grad_norm": 7.377877712249756, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 17.0262, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.002379712947125753, | |
| "grad_norm": 5.653985977172852, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 14.1928, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0024458160845459128, | |
| "grad_norm": 5.363378047943115, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 15.5034, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0025119192219660727, | |
| "grad_norm": 5.481878280639648, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 15.7485, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0025780223593862323, | |
| "grad_norm": 5.979918956756592, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 13.8195, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0025780223593862323, | |
| "eval_loss": NaN, | |
| "eval_runtime": 974.5754, | |
| "eval_samples_per_second": 26.144, | |
| "eval_steps_per_second": 13.072, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0026441254968063922, | |
| "grad_norm": 5.080392360687256, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 13.3521, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0027102286342265518, | |
| "grad_norm": 7.58221960067749, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 16.6197, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0027763317716467117, | |
| "grad_norm": 5.832557201385498, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 14.2864, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0028424349090668717, | |
| "grad_norm": 5.146307468414307, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 14.2749, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0029085380464870312, | |
| "grad_norm": 7.585702896118164, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 18.1122, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.002974641183907191, | |
| "grad_norm": 7.673133850097656, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 15.479, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.003040744321327351, | |
| "grad_norm": 5.815083980560303, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 12.2292, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0031068474587475107, | |
| "grad_norm": 5.892566204071045, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 13.7316, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0031729505961676707, | |
| "grad_norm": 5.821976184844971, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 14.5609, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0032390537335878302, | |
| "grad_norm": 5.67319393157959, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 16.7106, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.00330515687100799, | |
| "grad_norm": 6.600288391113281, | |
| "learning_rate": 0.0, | |
| "loss": 14.476, | |
| "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": 6.3324486107136e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |