| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.886165976524353, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.8569, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.9352289438247681, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.9157, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.6887379288673401, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8368, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5438244938850403, | |
| "learning_rate": 0.00019977668786231534, | |
| "loss": 0.7759, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.4049173593521118, | |
| "learning_rate": 0.000199107748815478, | |
| "loss": 0.7108, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.3374454379081726, | |
| "learning_rate": 0.0001979961705036587, | |
| "loss": 0.6949, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.31860846281051636, | |
| "learning_rate": 0.00019644691750543767, | |
| "loss": 0.6725, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.3116016089916229, | |
| "learning_rate": 0.0001944669091607919, | |
| "loss": 0.6814, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.3020254969596863, | |
| "learning_rate": 0.00019206498866764288, | |
| "loss": 0.6628, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.3751624524593353, | |
| "learning_rate": 0.00018925188358598813, | |
| "loss": 0.6491, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.33873409032821655, | |
| "learning_rate": 0.00018604015792601396, | |
| "loss": 0.5887, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3233942985534668, | |
| "learning_rate": 0.00018244415603417603, | |
| "loss": 0.5414, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.33184298872947693, | |
| "learning_rate": 0.0001784799385278661, | |
| "loss": 0.5534, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3251623511314392, | |
| "learning_rate": 0.00017416521056479577, | |
| "loss": 0.5303, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.30667614936828613, | |
| "learning_rate": 0.00016951924276746425, | |
| "loss": 0.5261, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.30262428522109985, | |
| "learning_rate": 0.00016456278515588024, | |
| "loss": 0.5101, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.30721279978752136, | |
| "learning_rate": 0.00015931797447293552, | |
| "loss": 0.5059, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.2939946949481964, | |
| "learning_rate": 0.00015380823531633729, | |
| "loss": 0.4949, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.28288447856903076, | |
| "learning_rate": 0.00014805817551866838, | |
| "loss": 0.4869, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2906552255153656, | |
| "learning_rate": 0.0001420934762428335, | |
| "loss": 0.4605, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.2544306218624115, | |
| "learning_rate": 0.00013594077728375128, | |
| "loss": 0.4436, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.2552658021450043, | |
| "learning_rate": 0.00012962755808856342, | |
| "loss": 0.4192, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.2546628713607788, | |
| "learning_rate": 0.00012318201502675285, | |
| "loss": 0.4106, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.23741765320301056, | |
| "learning_rate": 0.00011663293545831302, | |
| "loss": 0.3938, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.2525660991668701, | |
| "learning_rate": 0.00011000956916240985, | |
| "loss": 0.4017, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.2451285570859909, | |
| "learning_rate": 0.00010334149770076747, | |
| "loss": 0.3794, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.2524901032447815, | |
| "learning_rate": 9.665850229923258e-05, | |
| "loss": 0.3855, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.24502427875995636, | |
| "learning_rate": 8.999043083759017e-05, | |
| "loss": 0.3616, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.2539559602737427, | |
| "learning_rate": 8.336706454168701e-05, | |
| "loss": 0.3763, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.24636436998844147, | |
| "learning_rate": 7.681798497324716e-05, | |
| "loss": 0.3517, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.23345255851745605, | |
| "learning_rate": 7.037244191143661e-05, | |
| "loss": 0.3284, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.2510325014591217, | |
| "learning_rate": 6.405922271624874e-05, | |
| "loss": 0.3301, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 0.24276795983314514, | |
| "learning_rate": 5.790652375716652e-05, | |
| "loss": 0.3339, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.23698894679546356, | |
| "learning_rate": 5.1941824481331626e-05, | |
| "loss": 0.3273, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 0.24454469978809357, | |
| "learning_rate": 4.6191764683662744e-05, | |
| "loss": 0.3073, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.25007879734039307, | |
| "learning_rate": 4.0682025527064486e-05, | |
| "loss": 0.3149, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.25587430596351624, | |
| "learning_rate": 3.543721484411976e-05, | |
| "loss": 0.3081, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.2517322599887848, | |
| "learning_rate": 3.0480757232535772e-05, | |
| "loss": 0.3131, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 0.23721425235271454, | |
| "learning_rate": 2.5834789435204243e-05, | |
| "loss": 0.2987, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.23976309597492218, | |
| "learning_rate": 2.1520061472133902e-05, | |
| "loss": 0.3148, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 0.22691340744495392, | |
| "learning_rate": 1.7555843965823992e-05, | |
| "loss": 0.2777, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 0.23945631086826324, | |
| "learning_rate": 1.3959842073986085e-05, | |
| "loss": 0.2954, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 0.23303616046905518, | |
| "learning_rate": 1.0748116414011888e-05, | |
| "loss": 0.2882, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.24104540050029755, | |
| "learning_rate": 7.935011332357112e-06, | |
| "loss": 0.2939, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 0.23101571202278137, | |
| "learning_rate": 5.533090839208133e-06, | |
| "loss": 0.2789, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 0.2310316413640976, | |
| "learning_rate": 3.5530824945623542e-06, | |
| "loss": 0.2925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 0.23140206933021545, | |
| "learning_rate": 2.003829496341325e-06, | |
| "loss": 0.2824, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.22320057451725006, | |
| "learning_rate": 8.922511845219971e-07, | |
| "loss": 0.2975, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 0.22732405364513397, | |
| "learning_rate": 2.2331213768468363e-07, | |
| "loss": 0.2891, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.2486155927181244, | |
| "learning_rate": 0.0, | |
| "loss": 0.2815, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 7.457968683122688e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |