| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 5, | |
| "global_step": 30936, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0969743987587277, | |
| "grad_norm": 5.50258207321167, | |
| "learning_rate": 2.451704876061911e-05, | |
| "loss": 0.7801, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1939487975174554, | |
| "grad_norm": 2.8141658306121826, | |
| "learning_rate": 2.4032157958027852e-05, | |
| "loss": 0.4736, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29092319627618307, | |
| "grad_norm": 1.9647445678710938, | |
| "learning_rate": 2.3547267155436597e-05, | |
| "loss": 0.3846, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3878975950349108, | |
| "grad_norm": 0.9280088543891907, | |
| "learning_rate": 2.306334613445052e-05, | |
| "loss": 0.3573, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4848719937936385, | |
| "grad_norm": 4.660079479217529, | |
| "learning_rate": 2.2578455331859266e-05, | |
| "loss": 0.3135, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5818463925523661, | |
| "grad_norm": 2.9539144039154053, | |
| "learning_rate": 2.209356452926801e-05, | |
| "loss": 0.2992, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6788207913110939, | |
| "grad_norm": 1.4715936183929443, | |
| "learning_rate": 2.1608673726676753e-05, | |
| "loss": 0.2836, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7757951900698216, | |
| "grad_norm": 1.7886435985565186, | |
| "learning_rate": 2.1123782924085498e-05, | |
| "loss": 0.2698, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8727695888285493, | |
| "grad_norm": 1.0143980979919434, | |
| "learning_rate": 2.063889212149424e-05, | |
| "loss": 0.2549, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.969743987587277, | |
| "grad_norm": 4.67170524597168, | |
| "learning_rate": 2.0154001318902985e-05, | |
| "loss": 0.255, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.22126668691635132, | |
| "eval_runtime": 1558.3887, | |
| "eval_samples_per_second": 2.836, | |
| "eval_steps_per_second": 0.355, | |
| "step": 5156 | |
| }, | |
| { | |
| "epoch": 1.0667183863460046, | |
| "grad_norm": 1.17802894115448, | |
| "learning_rate": 1.966911051631173e-05, | |
| "loss": 0.198, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.1636927851047323, | |
| "grad_norm": 2.5160129070281982, | |
| "learning_rate": 1.9184219713720472e-05, | |
| "loss": 0.195, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.26066718386346, | |
| "grad_norm": 2.1824357509613037, | |
| "learning_rate": 1.8699328911129217e-05, | |
| "loss": 0.1864, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.3576415826221877, | |
| "grad_norm": 1.4108847379684448, | |
| "learning_rate": 1.821443810853796e-05, | |
| "loss": 0.1948, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.4546159813809154, | |
| "grad_norm": 1.5844694375991821, | |
| "learning_rate": 1.7730517087551886e-05, | |
| "loss": 0.1793, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.5515903801396431, | |
| "grad_norm": 0.8616577386856079, | |
| "learning_rate": 1.7245626284960627e-05, | |
| "loss": 0.1804, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.6485647788983708, | |
| "grad_norm": 2.6326045989990234, | |
| "learning_rate": 1.6760735482369372e-05, | |
| "loss": 0.1719, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.7455391776570985, | |
| "grad_norm": 1.3256356716156006, | |
| "learning_rate": 1.6275844679778114e-05, | |
| "loss": 0.1768, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.8425135764158262, | |
| "grad_norm": 2.5661003589630127, | |
| "learning_rate": 1.579192365879204e-05, | |
| "loss": 0.171, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.939487975174554, | |
| "grad_norm": 3.0119805335998535, | |
| "learning_rate": 1.5307032856200783e-05, | |
| "loss": 0.1654, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.16889868676662445, | |
| "eval_runtime": 1543.9946, | |
| "eval_samples_per_second": 2.863, | |
| "eval_steps_per_second": 0.358, | |
| "step": 10312 | |
| }, | |
| { | |
| "epoch": 2.0364623739332814, | |
| "grad_norm": 2.474855661392212, | |
| "learning_rate": 1.4825051398425075e-05, | |
| "loss": 0.1434, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.133436772692009, | |
| "grad_norm": 1.4645425081253052, | |
| "learning_rate": 1.4343069940649367e-05, | |
| "loss": 0.1296, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.230411171450737, | |
| "grad_norm": 1.1075003147125244, | |
| "learning_rate": 1.3858179138058112e-05, | |
| "loss": 0.1346, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.3273855702094646, | |
| "grad_norm": 2.3721115589141846, | |
| "learning_rate": 1.3373288335466855e-05, | |
| "loss": 0.1247, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.4243599689681923, | |
| "grad_norm": 0.8129625916481018, | |
| "learning_rate": 1.2888397532875595e-05, | |
| "loss": 0.1184, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.52133436772692, | |
| "grad_norm": 2.1738102436065674, | |
| "learning_rate": 1.240350673028434e-05, | |
| "loss": 0.1227, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.6183087664856477, | |
| "grad_norm": 2.0540452003479004, | |
| "learning_rate": 1.1918615927693085e-05, | |
| "loss": 0.1193, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.7152831652443754, | |
| "grad_norm": 1.0573948621749878, | |
| "learning_rate": 1.1433725125101828e-05, | |
| "loss": 0.1184, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.812257564003103, | |
| "grad_norm": 0.6909541487693787, | |
| "learning_rate": 1.094883432251057e-05, | |
| "loss": 0.1133, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.909231962761831, | |
| "grad_norm": 3.4285881519317627, | |
| "learning_rate": 1.0463943519919314e-05, | |
| "loss": 0.1221, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.1472880095243454, | |
| "eval_runtime": 1543.9214, | |
| "eval_samples_per_second": 2.863, | |
| "eval_steps_per_second": 0.358, | |
| "step": 15468 | |
| }, | |
| { | |
| "epoch": 3.0062063615205585, | |
| "grad_norm": 1.0781619548797607, | |
| "learning_rate": 9.979052717328059e-06, | |
| "loss": 0.1055, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.1031807602792862, | |
| "grad_norm": 1.153516411781311, | |
| "learning_rate": 9.494161914736802e-06, | |
| "loss": 0.0921, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.200155159038014, | |
| "grad_norm": 0.8690130114555359, | |
| "learning_rate": 9.009271112145545e-06, | |
| "loss": 0.0914, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.2971295577967417, | |
| "grad_norm": 0.9147763848304749, | |
| "learning_rate": 8.524380309554289e-06, | |
| "loss": 0.0955, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.3941039565554694, | |
| "grad_norm": 0.4283031225204468, | |
| "learning_rate": 8.039489506963032e-06, | |
| "loss": 0.0983, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.491078355314197, | |
| "grad_norm": 1.6501576900482178, | |
| "learning_rate": 7.554598704371776e-06, | |
| "loss": 0.0937, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.588052754072925, | |
| "grad_norm": 1.2180219888687134, | |
| "learning_rate": 7.069707901780519e-06, | |
| "loss": 0.0915, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.6850271528316525, | |
| "grad_norm": 1.027410864830017, | |
| "learning_rate": 6.584817099189262e-06, | |
| "loss": 0.091, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.78200155159038, | |
| "grad_norm": 1.701826810836792, | |
| "learning_rate": 6.100896078203189e-06, | |
| "loss": 0.0938, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.878975950349108, | |
| "grad_norm": 1.8526573181152344, | |
| "learning_rate": 5.616005275611932e-06, | |
| "loss": 0.0946, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.9759503491078356, | |
| "grad_norm": 2.527759075164795, | |
| "learning_rate": 5.131114473020676e-06, | |
| "loss": 0.0879, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.13828447461128235, | |
| "eval_runtime": 1543.3031, | |
| "eval_samples_per_second": 2.864, | |
| "eval_steps_per_second": 0.358, | |
| "step": 20624 | |
| }, | |
| { | |
| "epoch": 4.072924747866563, | |
| "grad_norm": 1.2997472286224365, | |
| "learning_rate": 1.482330921856514e-05, | |
| "loss": 0.0851, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.169899146625291, | |
| "grad_norm": 2.8086001873016357, | |
| "learning_rate": 1.4580868519560117e-05, | |
| "loss": 0.092, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.266873545384018, | |
| "grad_norm": 1.3651138544082642, | |
| "learning_rate": 1.4338427820555092e-05, | |
| "loss": 0.091, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.363847944142746, | |
| "grad_norm": 1.5494824647903442, | |
| "learning_rate": 1.409598712155007e-05, | |
| "loss": 0.0916, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.460822342901474, | |
| "grad_norm": 1.5494245290756226, | |
| "learning_rate": 1.3854031303943058e-05, | |
| "loss": 0.0919, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.557796741660201, | |
| "grad_norm": 2.636620044708252, | |
| "learning_rate": 1.3611590604938032e-05, | |
| "loss": 0.0936, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.654771140418929, | |
| "grad_norm": 1.1111247539520264, | |
| "learning_rate": 1.3369149905933009e-05, | |
| "loss": 0.0918, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.751745539177657, | |
| "grad_norm": 1.1152452230453491, | |
| "learning_rate": 1.3126709206927986e-05, | |
| "loss": 0.095, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.848719937936385, | |
| "grad_norm": 1.1507107019424438, | |
| "learning_rate": 1.2884268507922962e-05, | |
| "loss": 0.0933, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.945694336695112, | |
| "grad_norm": 0.9500592350959778, | |
| "learning_rate": 1.2641827808917939e-05, | |
| "loss": 0.0951, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.14213640987873077, | |
| "eval_runtime": 1544.0177, | |
| "eval_samples_per_second": 2.863, | |
| "eval_steps_per_second": 0.358, | |
| "step": 25780 | |
| }, | |
| { | |
| "epoch": 5.04266873545384, | |
| "grad_norm": 1.6337387561798096, | |
| "learning_rate": 1.2399871991310926e-05, | |
| "loss": 0.0852, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.139643134212568, | |
| "grad_norm": 0.6484285593032837, | |
| "learning_rate": 1.2157431292305903e-05, | |
| "loss": 0.0751, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 5.236617532971295, | |
| "grad_norm": 1.9428491592407227, | |
| "learning_rate": 1.1914990593300878e-05, | |
| "loss": 0.0789, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 5.333591931730023, | |
| "grad_norm": 1.7594366073608398, | |
| "learning_rate": 1.1672549894295855e-05, | |
| "loss": 0.0795, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 5.430566330488751, | |
| "grad_norm": 0.4171479046344757, | |
| "learning_rate": 1.1430109195290833e-05, | |
| "loss": 0.0775, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.5275407292474785, | |
| "grad_norm": 1.192670226097107, | |
| "learning_rate": 1.118766849628581e-05, | |
| "loss": 0.0822, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 5.624515128006206, | |
| "grad_norm": 1.016116976737976, | |
| "learning_rate": 1.0945227797280785e-05, | |
| "loss": 0.0811, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.721489526764934, | |
| "grad_norm": 0.925920307636261, | |
| "learning_rate": 1.0702787098275762e-05, | |
| "loss": 0.0767, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 5.818463925523662, | |
| "grad_norm": 1.05020010471344, | |
| "learning_rate": 1.046083128066875e-05, | |
| "loss": 0.0784, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.915438324282389, | |
| "grad_norm": 1.8532640933990479, | |
| "learning_rate": 1.0218390581663725e-05, | |
| "loss": 0.0777, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.14130142331123352, | |
| "eval_runtime": 1544.1253, | |
| "eval_samples_per_second": 2.862, | |
| "eval_steps_per_second": 0.358, | |
| "step": 30936 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 51560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2712794309494047e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |