|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 1605, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03115264797507788, |
|
"grad_norm": 57.86160659790039, |
|
"learning_rate": 9.976073288648913e-07, |
|
"loss": 0.7289, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.06230529595015576, |
|
"grad_norm": 60.39614486694336, |
|
"learning_rate": 9.904522149602086e-07, |
|
"loss": 0.6972, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.09345794392523364, |
|
"grad_norm": 151.33900451660156, |
|
"learning_rate": 9.78603137623984e-07, |
|
"loss": 0.5746, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.12461059190031153, |
|
"grad_norm": 52.87775802612305, |
|
"learning_rate": 9.621735006374983e-07, |
|
"loss": 0.4166, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1557632398753894, |
|
"grad_norm": 50.08223342895508, |
|
"learning_rate": 9.413205468734626e-07, |
|
"loss": 0.4558, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.18691588785046728, |
|
"grad_norm": 517.3528442382812, |
|
"learning_rate": 9.162438533740891e-07, |
|
"loss": 0.3246, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.21806853582554517, |
|
"grad_norm": 1556.693359375, |
|
"learning_rate": 8.871834212621772e-07, |
|
"loss": 0.4801, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.24922118380062305, |
|
"grad_norm": 16.822763442993164, |
|
"learning_rate": 8.544173787660787e-07, |
|
"loss": 0.2867, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.2803738317757009, |
|
"grad_norm": 2.0897529125213623, |
|
"learning_rate": 8.182593193421624e-07, |
|
"loss": 0.4674, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.3115264797507788, |
|
"grad_norm": 19.26500129699707, |
|
"learning_rate": 7.790553003707689e-07, |
|
"loss": 0.5003, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.3115264797507788, |
|
"eval_accuracy": 0.6294117647058823, |
|
"eval_loss": 2.713998317718506, |
|
"eval_runtime": 9.3954, |
|
"eval_samples_per_second": 36.188, |
|
"eval_steps_per_second": 9.047, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.3426791277258567, |
|
"grad_norm": 0.8728576898574829, |
|
"learning_rate": 7.371805311501904e-07, |
|
"loss": 0.5136, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.37383177570093457, |
|
"grad_norm": 167.8850860595703, |
|
"learning_rate": 6.930357818868407e-07, |
|
"loss": 0.2441, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.40498442367601245, |
|
"grad_norm": 177.43258666992188, |
|
"learning_rate": 6.470435480500362e-07, |
|
"loss": 0.4716, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 0.43613707165109034, |
|
"grad_norm": 289.399658203125, |
|
"learning_rate": 5.996440068011383e-07, |
|
"loss": 0.3897, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.4672897196261682, |
|
"grad_norm": 787.5913696289062, |
|
"learning_rate": 5.512908041968018e-07, |
|
"loss": 0.4029, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.4984423676012461, |
|
"grad_norm": 1.4492032960333745e-06, |
|
"learning_rate": 5.024467134856725e-07, |
|
"loss": 0.3285, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.5295950155763239, |
|
"grad_norm": 3.293706356544135e-07, |
|
"learning_rate": 4.5357920605161114e-07, |
|
"loss": 0.435, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 0.5607476635514018, |
|
"grad_norm": 1.7141993045806885, |
|
"learning_rate": 4.0515597739254617e-07, |
|
"loss": 0.3973, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.5919003115264797, |
|
"grad_norm": 2.922891617629375e-08, |
|
"learning_rate": 3.576404709544031e-07, |
|
"loss": 0.2163, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 0.6230529595015576, |
|
"grad_norm": 560.4365844726562, |
|
"learning_rate": 3.114874426600802e-07, |
|
"loss": 0.3939, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.6230529595015576, |
|
"eval_accuracy": 0.65, |
|
"eval_loss": 3.343003988265991, |
|
"eval_runtime": 8.9452, |
|
"eval_samples_per_second": 38.009, |
|
"eval_steps_per_second": 9.502, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.6542056074766355, |
|
"grad_norm": 0.0010744519531726837, |
|
"learning_rate": 2.671386085839682e-07, |
|
"loss": 0.2876, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 0.6853582554517134, |
|
"grad_norm": 2111.46875, |
|
"learning_rate": 2.2501841742674588e-07, |
|
"loss": 0.3456, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.7165109034267912, |
|
"grad_norm": 0.022361410781741142, |
|
"learning_rate": 1.855299882507616e-07, |
|
"loss": 0.3517, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 0.7476635514018691, |
|
"grad_norm": 2.9614204422045987e-08, |
|
"learning_rate": 1.490512523546559e-07, |
|
"loss": 0.4218, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.778816199376947, |
|
"grad_norm": 1.0811158546175648e-07, |
|
"learning_rate": 1.1593133621212453e-07, |
|
"loss": 0.2904, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.8099688473520249, |
|
"grad_norm": 1.012383222579956, |
|
"learning_rate": 8.648722009257315e-08, |
|
"loss": 0.4572, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.8411214953271028, |
|
"grad_norm": 3.705447113588889e-07, |
|
"learning_rate": 6.10007043429538e-08, |
|
"loss": 0.187, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 0.8722741433021807, |
|
"grad_norm": 9.330016837338917e-06, |
|
"learning_rate": 3.971571236554117e-08, |
|
"loss": 0.1274, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.9034267912772586, |
|
"grad_norm": 0.08515036106109619, |
|
"learning_rate": 2.2835956103996524e-08, |
|
"loss": 0.2892, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 0.9345794392523364, |
|
"grad_norm": 0.008006788790225983, |
|
"learning_rate": 1.0522986380618604e-08, |
|
"loss": 0.065, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.9345794392523364, |
|
"eval_accuracy": 0.6323529411764706, |
|
"eval_loss": 2.9739811420440674, |
|
"eval_runtime": 8.9019, |
|
"eval_samples_per_second": 38.194, |
|
"eval_steps_per_second": 9.549, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.9657320872274143, |
|
"grad_norm": 1.7190651533383061e-06, |
|
"learning_rate": 2.894646744385887e-09, |
|
"loss": 0.288, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 0.9968847352024922, |
|
"grad_norm": 909.2741088867188, |
|
"learning_rate": 2.3945627523891398e-11, |
|
"loss": 0.1994, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 1605, |
|
"total_flos": 0.0, |
|
"train_loss": 0.3749335529650852, |
|
"train_runtime": 2207.4532, |
|
"train_samples_per_second": 2.908, |
|
"train_steps_per_second": 0.727 |
|
} |
|
], |
|
"logging_steps": 50, |
|
"max_steps": 1605, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"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": 0.0, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|