| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 1000, |
| "global_step": 19132, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02613422538155969, |
| "grad_norm": 1.818863868713379, |
| "learning_rate": 4.8693288730922014e-05, |
| "loss": 4.3834, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.05226845076311938, |
| "grad_norm": 1.789357304573059, |
| "learning_rate": 4.738657746184403e-05, |
| "loss": 2.8977, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.05226845076311938, |
| "eval_accuracy": 0.4506061685839086, |
| "eval_loss": 2.428511142730713, |
| "eval_runtime": 54.4447, |
| "eval_samples_per_second": 113.069, |
| "eval_steps_per_second": 3.545, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.07840267614467906, |
| "grad_norm": 5.041560649871826, |
| "learning_rate": 4.607986619276605e-05, |
| "loss": 2.2708, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.10453690152623876, |
| "grad_norm": 1.5210151672363281, |
| "learning_rate": 4.477315492368807e-05, |
| "loss": 1.9691, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.10453690152623876, |
| "eval_accuracy": 0.5699825075886197, |
| "eval_loss": 1.7882978916168213, |
| "eval_runtime": 54.1381, |
| "eval_samples_per_second": 113.709, |
| "eval_steps_per_second": 3.565, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.13067112690779845, |
| "grad_norm": 2.0671706199645996, |
| "learning_rate": 4.346644365461008e-05, |
| "loss": 1.8168, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.15680535228935813, |
| "grad_norm": 1.402937412261963, |
| "learning_rate": 4.21597323855321e-05, |
| "loss": 1.7235, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.15680535228935813, |
| "eval_accuracy": 0.6021438684143834, |
| "eval_loss": 1.6068116426467896, |
| "eval_runtime": 53.8571, |
| "eval_samples_per_second": 114.303, |
| "eval_steps_per_second": 3.584, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.18293957767091784, |
| "grad_norm": 1.6213388442993164, |
| "learning_rate": 4.085302111645411e-05, |
| "loss": 1.6543, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.20907380305247752, |
| "grad_norm": 1.2356746196746826, |
| "learning_rate": 3.954630984737613e-05, |
| "loss": 1.6024, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.20907380305247752, |
| "eval_accuracy": 0.6207478164656056, |
| "eval_loss": 1.5058479309082031, |
| "eval_runtime": 53.7796, |
| "eval_samples_per_second": 114.467, |
| "eval_steps_per_second": 3.589, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.2352080284340372, |
| "grad_norm": 1.2803748846054077, |
| "learning_rate": 3.823959857829814e-05, |
| "loss": 1.5632, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.2613422538155969, |
| "grad_norm": 1.2122167348861694, |
| "learning_rate": 3.6932887309220156e-05, |
| "loss": 1.5325, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.2613422538155969, |
| "eval_accuracy": 0.6314862134518803, |
| "eval_loss": 1.4471280574798584, |
| "eval_runtime": 54.1245, |
| "eval_samples_per_second": 113.738, |
| "eval_steps_per_second": 3.566, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.2874764791971566, |
| "grad_norm": 1.2028292417526245, |
| "learning_rate": 3.5626176040142174e-05, |
| "loss": 1.5035, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.31361070457871626, |
| "grad_norm": 1.2411212921142578, |
| "learning_rate": 3.4319464771064186e-05, |
| "loss": 1.481, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.31361070457871626, |
| "eval_accuracy": 0.6399610454034148, |
| "eval_loss": 1.403512954711914, |
| "eval_runtime": 54.3345, |
| "eval_samples_per_second": 113.298, |
| "eval_steps_per_second": 3.552, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.339744929960276, |
| "grad_norm": 1.1803300380706787, |
| "learning_rate": 3.3012753501986204e-05, |
| "loss": 1.4594, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.3658791553418357, |
| "grad_norm": 1.2041865587234497, |
| "learning_rate": 3.1706042232908215e-05, |
| "loss": 1.4382, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.3658791553418357, |
| "eval_accuracy": 0.646798107465906, |
| "eval_loss": 1.3690356016159058, |
| "eval_runtime": 54.4547, |
| "eval_samples_per_second": 113.048, |
| "eval_steps_per_second": 3.544, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.39201338072339537, |
| "grad_norm": 1.1233160495758057, |
| "learning_rate": 3.0399330963830237e-05, |
| "loss": 1.4224, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.41814760610495505, |
| "grad_norm": 1.1373196840286255, |
| "learning_rate": 2.9092619694752248e-05, |
| "loss": 1.4083, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.41814760610495505, |
| "eval_accuracy": 0.6527462577736111, |
| "eval_loss": 1.337209939956665, |
| "eval_runtime": 54.1548, |
| "eval_samples_per_second": 113.674, |
| "eval_steps_per_second": 3.564, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.44428183148651473, |
| "grad_norm": 1.1343424320220947, |
| "learning_rate": 2.7785908425674263e-05, |
| "loss": 1.3984, |
| "step": 8500 |
| }, |
| { |
| "epoch": 0.4704160568680744, |
| "grad_norm": 1.1359299421310425, |
| "learning_rate": 2.647919715659628e-05, |
| "loss": 1.3835, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.4704160568680744, |
| "eval_accuracy": 0.6568230567004383, |
| "eval_loss": 1.3156871795654297, |
| "eval_runtime": 54.1248, |
| "eval_samples_per_second": 113.737, |
| "eval_steps_per_second": 3.566, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.4965502822496341, |
| "grad_norm": 1.1124554872512817, |
| "learning_rate": 2.5172485887518292e-05, |
| "loss": 1.3722, |
| "step": 9500 |
| }, |
| { |
| "epoch": 0.5226845076311938, |
| "grad_norm": 1.170714259147644, |
| "learning_rate": 2.3865774618440313e-05, |
| "loss": 1.3618, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5226845076311938, |
| "eval_accuracy": 0.6607685355091505, |
| "eval_loss": 1.2963454723358154, |
| "eval_runtime": 54.2271, |
| "eval_samples_per_second": 113.522, |
| "eval_steps_per_second": 3.559, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5488187330127535, |
| "grad_norm": 1.0964117050170898, |
| "learning_rate": 2.2559063349362325e-05, |
| "loss": 1.3532, |
| "step": 10500 |
| }, |
| { |
| "epoch": 0.5749529583943132, |
| "grad_norm": 1.0944701433181763, |
| "learning_rate": 2.125235208028434e-05, |
| "loss": 1.3439, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.5749529583943132, |
| "eval_accuracy": 0.6641806355068004, |
| "eval_loss": 1.2785439491271973, |
| "eval_runtime": 54.0593, |
| "eval_samples_per_second": 113.875, |
| "eval_steps_per_second": 3.57, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.6010871837758729, |
| "grad_norm": 1.1631088256835938, |
| "learning_rate": 1.9945640811206358e-05, |
| "loss": 1.3347, |
| "step": 11500 |
| }, |
| { |
| "epoch": 0.6272214091574325, |
| "grad_norm": 1.0688389539718628, |
| "learning_rate": 1.8638929542128372e-05, |
| "loss": 1.3279, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6272214091574325, |
| "eval_accuracy": 0.6668480059349706, |
| "eval_loss": 1.2657736539840698, |
| "eval_runtime": 54.0951, |
| "eval_samples_per_second": 113.8, |
| "eval_steps_per_second": 3.568, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6533556345389923, |
| "grad_norm": 1.103243112564087, |
| "learning_rate": 1.7332218273050387e-05, |
| "loss": 1.324, |
| "step": 12500 |
| }, |
| { |
| "epoch": 0.679489859920552, |
| "grad_norm": 1.1131521463394165, |
| "learning_rate": 1.6025507003972405e-05, |
| "loss": 1.3158, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.679489859920552, |
| "eval_accuracy": 0.6693424530153449, |
| "eval_loss": 1.2527960538864136, |
| "eval_runtime": 53.9587, |
| "eval_samples_per_second": 114.087, |
| "eval_steps_per_second": 3.577, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.7056240853021116, |
| "grad_norm": 1.1235332489013672, |
| "learning_rate": 1.471879573489442e-05, |
| "loss": 1.3089, |
| "step": 13500 |
| }, |
| { |
| "epoch": 0.7317583106836714, |
| "grad_norm": 1.1117353439331055, |
| "learning_rate": 1.3412084465816434e-05, |
| "loss": 1.3022, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.7317583106836714, |
| "eval_accuracy": 0.671401654093599, |
| "eval_loss": 1.2427349090576172, |
| "eval_runtime": 52.9223, |
| "eval_samples_per_second": 116.322, |
| "eval_steps_per_second": 3.647, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.757892536065231, |
| "grad_norm": 1.105966567993164, |
| "learning_rate": 1.210537319673845e-05, |
| "loss": 1.2978, |
| "step": 14500 |
| }, |
| { |
| "epoch": 0.7840267614467907, |
| "grad_norm": 1.1137796640396118, |
| "learning_rate": 1.0798661927660466e-05, |
| "loss": 1.295, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.7840267614467907, |
| "eval_accuracy": 0.6732677653730286, |
| "eval_loss": 1.2337384223937988, |
| "eval_runtime": 52.9824, |
| "eval_samples_per_second": 116.189, |
| "eval_steps_per_second": 3.643, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.8101609868283504, |
| "grad_norm": 1.1172099113464355, |
| "learning_rate": 9.49195065858248e-06, |
| "loss": 1.2892, |
| "step": 15500 |
| }, |
| { |
| "epoch": 0.8362952122099101, |
| "grad_norm": 1.1058783531188965, |
| "learning_rate": 8.185239389504495e-06, |
| "loss": 1.285, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8362952122099101, |
| "eval_accuracy": 0.6751054530718745, |
| "eval_loss": 1.2253085374832153, |
| "eval_runtime": 52.8664, |
| "eval_samples_per_second": 116.445, |
| "eval_steps_per_second": 3.651, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8624294375914698, |
| "grad_norm": 1.0990358591079712, |
| "learning_rate": 6.878528120426511e-06, |
| "loss": 1.2792, |
| "step": 16500 |
| }, |
| { |
| "epoch": 0.8885636629730295, |
| "grad_norm": 1.103635549545288, |
| "learning_rate": 5.571816851348526e-06, |
| "loss": 1.2774, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.8885636629730295, |
| "eval_accuracy": 0.676485822826136, |
| "eval_loss": 1.2190231084823608, |
| "eval_runtime": 52.4126, |
| "eval_samples_per_second": 117.453, |
| "eval_steps_per_second": 3.682, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.9146978883545892, |
| "grad_norm": 1.1323543787002563, |
| "learning_rate": 4.265105582270542e-06, |
| "loss": 1.2731, |
| "step": 17500 |
| }, |
| { |
| "epoch": 0.9408321137361488, |
| "grad_norm": 1.110835313796997, |
| "learning_rate": 2.958394313192557e-06, |
| "loss": 1.2734, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9408321137361488, |
| "eval_accuracy": 0.6775752875545368, |
| "eval_loss": 1.2133620977401733, |
| "eval_runtime": 52.4851, |
| "eval_samples_per_second": 117.29, |
| "eval_steps_per_second": 3.677, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9669663391177086, |
| "grad_norm": 1.1150257587432861, |
| "learning_rate": 1.6516830441145725e-06, |
| "loss": 1.2682, |
| "step": 18500 |
| }, |
| { |
| "epoch": 0.9931005644992682, |
| "grad_norm": 1.1233183145523071, |
| "learning_rate": 3.4497177503658794e-07, |
| "loss": 1.2705, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.9931005644992682, |
| "eval_accuracy": 0.6783100132939786, |
| "eval_loss": 1.2098151445388794, |
| "eval_runtime": 52.3265, |
| "eval_samples_per_second": 117.646, |
| "eval_steps_per_second": 3.688, |
| "step": 19000 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 19132, |
| "total_flos": 3.19927531732992e+17, |
| "train_loss": 1.5522811494062019, |
| "train_runtime": 7453.1649, |
| "train_samples_per_second": 82.14, |
| "train_steps_per_second": 2.567 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 19132, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 1000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.19927531732992e+17, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
