| { |
| "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": 1.0069942474365234, |
| "learning_rate": 4.8693288730922014e-05, |
| "loss": 4.3789, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.05226845076311938, |
| "grad_norm": 1.8369040489196777, |
| "learning_rate": 4.738657746184403e-05, |
| "loss": 2.9037, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.05226845076311938, |
| "eval_accuracy": 0.4486757469685219, |
| "eval_loss": 2.4443750381469727, |
| "eval_runtime": 52.7365, |
| "eval_samples_per_second": 116.731, |
| "eval_steps_per_second": 3.66, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.07840267614467906, |
| "grad_norm": 5.2621636390686035, |
| "learning_rate": 4.607986619276605e-05, |
| "loss": 2.2813, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.10453690152623876, |
| "grad_norm": 1.531705617904663, |
| "learning_rate": 4.477315492368807e-05, |
| "loss": 1.9735, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.10453690152623876, |
| "eval_accuracy": 0.5685718087623388, |
| "eval_loss": 1.796225905418396, |
| "eval_runtime": 52.7973, |
| "eval_samples_per_second": 116.597, |
| "eval_steps_per_second": 3.655, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.13067112690779845, |
| "grad_norm": 1.563498854637146, |
| "learning_rate": 4.346644365461008e-05, |
| "loss": 1.8191, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.15680535228935813, |
| "grad_norm": 1.4150679111480713, |
| "learning_rate": 4.21597323855321e-05, |
| "loss": 1.7239, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.15680535228935813, |
| "eval_accuracy": 0.6016185879419231, |
| "eval_loss": 1.6086262464523315, |
| "eval_runtime": 52.8364, |
| "eval_samples_per_second": 116.511, |
| "eval_steps_per_second": 3.653, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.18293957767091784, |
| "grad_norm": 1.4226484298706055, |
| "learning_rate": 4.085302111645411e-05, |
| "loss": 1.6536, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.20907380305247752, |
| "grad_norm": 1.2391685247421265, |
| "learning_rate": 3.954630984737613e-05, |
| "loss": 1.6013, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.20907380305247752, |
| "eval_accuracy": 0.6210625401344134, |
| "eval_loss": 1.50575852394104, |
| "eval_runtime": 52.7674, |
| "eval_samples_per_second": 116.663, |
| "eval_steps_per_second": 3.658, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.2352080284340372, |
| "grad_norm": 1.4180678129196167, |
| "learning_rate": 3.823959857829814e-05, |
| "loss": 1.5616, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.2613422538155969, |
| "grad_norm": 1.303842306137085, |
| "learning_rate": 3.6932887309220156e-05, |
| "loss": 1.5297, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.2613422538155969, |
| "eval_accuracy": 0.6315313100825268, |
| "eval_loss": 1.446053385734558, |
| "eval_runtime": 52.6764, |
| "eval_samples_per_second": 116.865, |
| "eval_steps_per_second": 3.664, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.2874764791971566, |
| "grad_norm": 1.2424602508544922, |
| "learning_rate": 3.5626176040142174e-05, |
| "loss": 1.5007, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.31361070457871626, |
| "grad_norm": 1.2661899328231812, |
| "learning_rate": 3.4319464771064186e-05, |
| "loss": 1.478, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.31361070457871626, |
| "eval_accuracy": 0.640106656707298, |
| "eval_loss": 1.4016118049621582, |
| "eval_runtime": 52.5005, |
| "eval_samples_per_second": 117.256, |
| "eval_steps_per_second": 3.676, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.339744929960276, |
| "grad_norm": 1.1755191087722778, |
| "learning_rate": 3.3012753501986204e-05, |
| "loss": 1.4558, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.3658791553418357, |
| "grad_norm": 1.3452744483947754, |
| "learning_rate": 3.1706042232908215e-05, |
| "loss": 1.4343, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.3658791553418357, |
| "eval_accuracy": 0.6475163189462378, |
| "eval_loss": 1.3634966611862183, |
| "eval_runtime": 52.4288, |
| "eval_samples_per_second": 117.416, |
| "eval_steps_per_second": 3.681, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.39201338072339537, |
| "grad_norm": 1.1488875150680542, |
| "learning_rate": 3.0399330963830237e-05, |
| "loss": 1.4183, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.41814760610495505, |
| "grad_norm": 1.1783889532089233, |
| "learning_rate": 2.9092619694752248e-05, |
| "loss": 1.4042, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.41814760610495505, |
| "eval_accuracy": 0.6532186608587288, |
| "eval_loss": 1.3342305421829224, |
| "eval_runtime": 52.5027, |
| "eval_samples_per_second": 117.251, |
| "eval_steps_per_second": 3.676, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.44428183148651473, |
| "grad_norm": 1.1511882543563843, |
| "learning_rate": 2.7785908425674263e-05, |
| "loss": 1.3937, |
| "step": 8500 |
| }, |
| { |
| "epoch": 0.4704160568680744, |
| "grad_norm": 1.1524832248687744, |
| "learning_rate": 2.647919715659628e-05, |
| "loss": 1.3786, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.4704160568680744, |
| "eval_accuracy": 0.6576487696559381, |
| "eval_loss": 1.3107670545578003, |
| "eval_runtime": 53.6197, |
| "eval_samples_per_second": 114.808, |
| "eval_steps_per_second": 3.599, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.4965502822496341, |
| "grad_norm": 1.1211012601852417, |
| "learning_rate": 2.5172485887518292e-05, |
| "loss": 1.3674, |
| "step": 9500 |
| }, |
| { |
| "epoch": 0.5226845076311938, |
| "grad_norm": 1.1335201263427734, |
| "learning_rate": 2.3865774618440313e-05, |
| "loss": 1.3566, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5226845076311938, |
| "eval_accuracy": 0.6616726911954227, |
| "eval_loss": 1.2912213802337646, |
| "eval_runtime": 53.4337, |
| "eval_samples_per_second": 115.208, |
| "eval_steps_per_second": 3.612, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.5488187330127535, |
| "grad_norm": 1.109776496887207, |
| "learning_rate": 2.2559063349362325e-05, |
| "loss": 1.3481, |
| "step": 10500 |
| }, |
| { |
| "epoch": 0.5749529583943132, |
| "grad_norm": 1.0994426012039185, |
| "learning_rate": 2.125235208028434e-05, |
| "loss": 1.3389, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.5749529583943132, |
| "eval_accuracy": 0.665065101114903, |
| "eval_loss": 1.273192286491394, |
| "eval_runtime": 53.8053, |
| "eval_samples_per_second": 114.413, |
| "eval_steps_per_second": 3.587, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.6010871837758729, |
| "grad_norm": 1.1425756216049194, |
| "learning_rate": 1.9945640811206358e-05, |
| "loss": 1.3294, |
| "step": 11500 |
| }, |
| { |
| "epoch": 0.6272214091574325, |
| "grad_norm": 1.1189128160476685, |
| "learning_rate": 1.8638929542128372e-05, |
| "loss": 1.3228, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6272214091574325, |
| "eval_accuracy": 0.6674850752383293, |
| "eval_loss": 1.2623748779296875, |
| "eval_runtime": 53.8318, |
| "eval_samples_per_second": 114.356, |
| "eval_steps_per_second": 3.585, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.6533556345389923, |
| "grad_norm": 1.1318820714950562, |
| "learning_rate": 1.7332218273050387e-05, |
| "loss": 1.3188, |
| "step": 12500 |
| }, |
| { |
| "epoch": 0.679489859920552, |
| "grad_norm": 1.1322083473205566, |
| "learning_rate": 1.6025507003972405e-05, |
| "loss": 1.3105, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.679489859920552, |
| "eval_accuracy": 0.6702737619545769, |
| "eval_loss": 1.2479877471923828, |
| "eval_runtime": 53.2542, |
| "eval_samples_per_second": 115.596, |
| "eval_steps_per_second": 3.624, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.7056240853021116, |
| "grad_norm": 1.105891227722168, |
| "learning_rate": 1.471879573489442e-05, |
| "loss": 1.3034, |
| "step": 13500 |
| }, |
| { |
| "epoch": 0.7317583106836714, |
| "grad_norm": 1.1217329502105713, |
| "learning_rate": 1.3412084465816434e-05, |
| "loss": 1.2968, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.7317583106836714, |
| "eval_accuracy": 0.6723256586489939, |
| "eval_loss": 1.237587809562683, |
| "eval_runtime": 52.8726, |
| "eval_samples_per_second": 116.431, |
| "eval_steps_per_second": 3.65, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.757892536065231, |
| "grad_norm": 1.0946693420410156, |
| "learning_rate": 1.210537319673845e-05, |
| "loss": 1.2924, |
| "step": 14500 |
| }, |
| { |
| "epoch": 0.7840267614467907, |
| "grad_norm": 1.081743836402893, |
| "learning_rate": 1.0798661927660466e-05, |
| "loss": 1.2894, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.7840267614467907, |
| "eval_accuracy": 0.674236548977164, |
| "eval_loss": 1.2293522357940674, |
| "eval_runtime": 53.025, |
| "eval_samples_per_second": 116.096, |
| "eval_steps_per_second": 3.64, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.8101609868283504, |
| "grad_norm": 1.1411429643630981, |
| "learning_rate": 9.49195065858248e-06, |
| "loss": 1.2839, |
| "step": 15500 |
| }, |
| { |
| "epoch": 0.8362952122099101, |
| "grad_norm": 1.1418683528900146, |
| "learning_rate": 8.185239389504495e-06, |
| "loss": 1.2795, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8362952122099101, |
| "eval_accuracy": 0.6758117552307328, |
| "eval_loss": 1.2207015752792358, |
| "eval_runtime": 53.0931, |
| "eval_samples_per_second": 115.947, |
| "eval_steps_per_second": 3.635, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.8624294375914698, |
| "grad_norm": 1.115324854850769, |
| "learning_rate": 6.878528120426511e-06, |
| "loss": 1.2737, |
| "step": 16500 |
| }, |
| { |
| "epoch": 0.8885636629730295, |
| "grad_norm": 1.1184618473052979, |
| "learning_rate": 5.571816851348526e-06, |
| "loss": 1.2718, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.8885636629730295, |
| "eval_accuracy": 0.6774212603301455, |
| "eval_loss": 1.2131068706512451, |
| "eval_runtime": 53.0358, |
| "eval_samples_per_second": 116.073, |
| "eval_steps_per_second": 3.639, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.9146978883545892, |
| "grad_norm": 1.1303164958953857, |
| "learning_rate": 4.265105582270542e-06, |
| "loss": 1.2676, |
| "step": 17500 |
| }, |
| { |
| "epoch": 0.9408321137361488, |
| "grad_norm": 1.0928161144256592, |
| "learning_rate": 2.958394313192557e-06, |
| "loss": 1.2679, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9408321137361488, |
| "eval_accuracy": 0.6784538778973791, |
| "eval_loss": 1.2084242105484009, |
| "eval_runtime": 52.7134, |
| "eval_samples_per_second": 116.782, |
| "eval_steps_per_second": 3.661, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.9669663391177086, |
| "grad_norm": 1.1279501914978027, |
| "learning_rate": 1.6516830441145725e-06, |
| "loss": 1.2626, |
| "step": 18500 |
| }, |
| { |
| "epoch": 0.9931005644992682, |
| "grad_norm": 1.1036733388900757, |
| "learning_rate": 3.4497177503658794e-07, |
| "loss": 1.2646, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.9931005644992682, |
| "eval_accuracy": 0.6793310391216447, |
| "eval_loss": 1.204505443572998, |
| "eval_runtime": 52.9636, |
| "eval_samples_per_second": 116.231, |
| "eval_steps_per_second": 3.644, |
| "step": 19000 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 19132, |
| "total_flos": 3.19927531732992e+17, |
| "train_loss": 1.5489698861050551, |
| "train_runtime": 7449.0972, |
| "train_samples_per_second": 82.185, |
| "train_steps_per_second": 2.568 |
| } |
| ], |
| "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 |
| } |
|
|
