| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03931976801336872, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007863953602673744, | |
| "grad_norm": 1.727420449256897, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2759, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007863953602673744, | |
| "eval_loss": 2.9491188526153564, | |
| "eval_runtime": 802.2962, | |
| "eval_samples_per_second": 2.67, | |
| "eval_steps_per_second": 1.335, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015727907205347487, | |
| "grad_norm": 2.06775164604187, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2877, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023591860808021233, | |
| "grad_norm": 2.407663583755493, | |
| "learning_rate": 3e-05, | |
| "loss": 1.367, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031455814410694975, | |
| "grad_norm": 1.9768824577331543, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2932, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003931976801336872, | |
| "grad_norm": 2.2676198482513428, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4012, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004718372161604247, | |
| "grad_norm": 2.2389116287231445, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4455, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005504767521871621, | |
| "grad_norm": 3.368894577026367, | |
| "learning_rate": 7e-05, | |
| "loss": 1.6058, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006291162882138995, | |
| "grad_norm": 2.1887245178222656, | |
| "learning_rate": 8e-05, | |
| "loss": 1.315, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00707755824240637, | |
| "grad_norm": 2.749788284301758, | |
| "learning_rate": 9e-05, | |
| "loss": 1.4487, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007863953602673744, | |
| "grad_norm": 2.44626784324646, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9548, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008650348962941118, | |
| "grad_norm": 2.37338924407959, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.8456, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009436744323208493, | |
| "grad_norm": 1.9337718486785889, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.8131, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010223139683475867, | |
| "grad_norm": 5.305741786956787, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.3641, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010223139683475867, | |
| "eval_loss": 1.1911835670471191, | |
| "eval_runtime": 790.9924, | |
| "eval_samples_per_second": 2.708, | |
| "eval_steps_per_second": 1.354, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011009535043743243, | |
| "grad_norm": 1.8393322229385376, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8994, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011795930404010616, | |
| "grad_norm": 2.288557529449463, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.964, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01258232576427799, | |
| "grad_norm": 2.239872455596924, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.8643, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013368721124545365, | |
| "grad_norm": 2.4298009872436523, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.9466, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01415511648481274, | |
| "grad_norm": 3.0432651042938232, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0756, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014941511845080115, | |
| "grad_norm": 1.3065857887268066, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.6739, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.01572790720534749, | |
| "grad_norm": 2.0720527172088623, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.8697, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.016514302565614862, | |
| "grad_norm": 1.8250558376312256, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.7318, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017300697925882236, | |
| "grad_norm": 2.166428565979004, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.7315, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.018087093286149613, | |
| "grad_norm": 2.2534689903259277, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.9042, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.018873488646416987, | |
| "grad_norm": 1.4026483297348022, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.7183, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01965988400668436, | |
| "grad_norm": 1.6705015897750854, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.6787, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.020446279366951734, | |
| "grad_norm": 2.1395044326782227, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.9407, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.020446279366951734, | |
| "eval_loss": 0.8785327672958374, | |
| "eval_runtime": 781.2326, | |
| "eval_samples_per_second": 2.742, | |
| "eval_steps_per_second": 1.371, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021232674727219108, | |
| "grad_norm": 1.2242342233657837, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.6702, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.022019070087486485, | |
| "grad_norm": 1.7602314949035645, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.9246, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02280546544775386, | |
| "grad_norm": 2.0709216594696045, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.8103, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.023591860808021232, | |
| "grad_norm": 1.5635908842086792, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8027, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.024378256168288606, | |
| "grad_norm": 1.6824203729629517, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.776, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.02516465152855598, | |
| "grad_norm": 1.457542061805725, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.5255, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.025951046888823357, | |
| "grad_norm": 1.5966757535934448, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.6032, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02673744224909073, | |
| "grad_norm": 1.5422358512878418, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.548, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.027523837609358105, | |
| "grad_norm": 1.623889446258545, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.7702, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02831023296962548, | |
| "grad_norm": 1.665121078491211, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.6063, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.029096628329892852, | |
| "grad_norm": 1.3971141576766968, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.48, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.02988302369016023, | |
| "grad_norm": 1.3747204542160034, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.6483, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.030669419050427603, | |
| "grad_norm": 1.3398278951644897, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.7969, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.030669419050427603, | |
| "eval_loss": 0.8240298628807068, | |
| "eval_runtime": 792.48, | |
| "eval_samples_per_second": 2.703, | |
| "eval_steps_per_second": 1.351, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03145581441069498, | |
| "grad_norm": 1.3485429286956787, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.6484, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03224220977096235, | |
| "grad_norm": 1.2892612218856812, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.7116, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.033028605131229724, | |
| "grad_norm": 1.8880107402801514, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.5925, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0338150004914971, | |
| "grad_norm": 1.4518492221832275, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.6824, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03460139585176447, | |
| "grad_norm": 2.3983242511749268, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.9552, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03538779121203185, | |
| "grad_norm": 1.543707013130188, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.8833, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.036174186572299226, | |
| "grad_norm": 1.7945361137390137, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.8763, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0369605819325666, | |
| "grad_norm": 1.6415727138519287, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.7373, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03774697729283397, | |
| "grad_norm": 1.7929890155792236, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.6606, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03853337265310135, | |
| "grad_norm": 1.3461647033691406, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.7808, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03931976801336872, | |
| "grad_norm": 1.1689928770065308, | |
| "learning_rate": 0.0, | |
| "loss": 0.7459, | |
| "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": 3.067577446170624e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |