| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.010301490282260834, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001373532037634778, | |
| "grad_norm": 3.1040570735931396, | |
| "learning_rate": 1e-05, | |
| "loss": 4.1873, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001373532037634778, | |
| "eval_loss": 5.127926349639893, | |
| "eval_runtime": 1387.7989, | |
| "eval_samples_per_second": 4.418, | |
| "eval_steps_per_second": 0.553, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002747064075269556, | |
| "grad_norm": 5.985080718994141, | |
| "learning_rate": 2e-05, | |
| "loss": 5.9119, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00041205961129043336, | |
| "grad_norm": 4.335269927978516, | |
| "learning_rate": 3e-05, | |
| "loss": 5.4319, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005494128150539112, | |
| "grad_norm": 3.2608160972595215, | |
| "learning_rate": 4e-05, | |
| "loss": 4.3406, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.000686766018817389, | |
| "grad_norm": 3.8993234634399414, | |
| "learning_rate": 5e-05, | |
| "loss": 4.6659, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008241192225808667, | |
| "grad_norm": 4.482377529144287, | |
| "learning_rate": 6e-05, | |
| "loss": 4.7347, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0009614724263443445, | |
| "grad_norm": 4.670428276062012, | |
| "learning_rate": 7e-05, | |
| "loss": 4.4137, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0010988256301078224, | |
| "grad_norm": 5.127176284790039, | |
| "learning_rate": 8e-05, | |
| "loss": 4.4507, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0012361788338713, | |
| "grad_norm": 6.486532688140869, | |
| "learning_rate": 9e-05, | |
| "loss": 4.733, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0012361788338713, | |
| "eval_loss": 3.352243185043335, | |
| "eval_runtime": 1387.8729, | |
| "eval_samples_per_second": 4.418, | |
| "eval_steps_per_second": 0.553, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001373532037634778, | |
| "grad_norm": 5.012729167938232, | |
| "learning_rate": 0.0001, | |
| "loss": 3.3681, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0015108852413982556, | |
| "grad_norm": 3.187636613845825, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.6068, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0016482384451617334, | |
| "grad_norm": 4.54232931137085, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.7261, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001785591648925211, | |
| "grad_norm": 4.337673664093018, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.4543, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.001922944852688689, | |
| "grad_norm": 4.798433303833008, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.211, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0020602980564521666, | |
| "grad_norm": 3.4199917316436768, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.8941, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0021976512602156447, | |
| "grad_norm": 4.503613471984863, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.8994, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0023350044639791224, | |
| "grad_norm": 2.812070608139038, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 1.5397, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0024723576677426, | |
| "grad_norm": 3.481353282928467, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.48, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0024723576677426, | |
| "eval_loss": 1.573837161064148, | |
| "eval_runtime": 1387.6231, | |
| "eval_samples_per_second": 4.418, | |
| "eval_steps_per_second": 0.553, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0026097108715060777, | |
| "grad_norm": 2.8117356300354004, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.4456, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.002747064075269556, | |
| "grad_norm": 3.217996835708618, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.6385, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0028844172790330335, | |
| "grad_norm": 2.422680377960205, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.5465, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.003021770482796511, | |
| "grad_norm": 2.074409246444702, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 1.1328, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.003159123686559989, | |
| "grad_norm": 2.0114176273345947, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.1746, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.003296476890323467, | |
| "grad_norm": 1.7757854461669922, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.0242, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0034338300940869445, | |
| "grad_norm": 2.0996131896972656, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.9594, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.003571183297850422, | |
| "grad_norm": 1.5292807817459106, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.7581, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0037085365016139003, | |
| "grad_norm": 2.0904600620269775, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.0291, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0037085365016139003, | |
| "eval_loss": 1.0520473718643188, | |
| "eval_runtime": 1387.7815, | |
| "eval_samples_per_second": 4.418, | |
| "eval_steps_per_second": 0.553, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.003845889705377378, | |
| "grad_norm": 2.166792631149292, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0476, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.003983242909140856, | |
| "grad_norm": 4.157371997833252, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.1389, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.004120596112904333, | |
| "grad_norm": 2.744110345840454, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.005, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.004257949316667811, | |
| "grad_norm": 2.2463557720184326, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.7509, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0043953025204312894, | |
| "grad_norm": 2.687026023864746, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.8431, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.004532655724194767, | |
| "grad_norm": 2.2226080894470215, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.9746, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.004670008927958245, | |
| "grad_norm": 1.9216952323913574, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.6121, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.004807362131721722, | |
| "grad_norm": 2.6237952709198, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.6943, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0049447153354852, | |
| "grad_norm": 1.6128216981887817, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.5205, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0049447153354852, | |
| "eval_loss": 0.6823018193244934, | |
| "eval_runtime": 1386.5683, | |
| "eval_samples_per_second": 4.422, | |
| "eval_steps_per_second": 0.553, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.005082068539248678, | |
| "grad_norm": 2.30482816696167, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.4894, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.005219421743012155, | |
| "grad_norm": 4.286962985992432, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6738, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0053567749467756335, | |
| "grad_norm": 3.888533353805542, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.4983, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.005494128150539112, | |
| "grad_norm": 2.3330461978912354, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.4765, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.005631481354302589, | |
| "grad_norm": 4.286584377288818, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.7936, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.005768834558066067, | |
| "grad_norm": 3.957209587097168, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.7258, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.005906187761829545, | |
| "grad_norm": 5.282871723175049, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.5848, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.006043540965593022, | |
| "grad_norm": 2.5516767501831055, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.397, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0061808941693565, | |
| "grad_norm": 2.4859097003936768, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.4567, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0061808941693565, | |
| "eval_loss": 0.404715359210968, | |
| "eval_runtime": 1387.0837, | |
| "eval_samples_per_second": 4.42, | |
| "eval_steps_per_second": 0.553, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.006318247373119978, | |
| "grad_norm": 2.7337472438812256, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.4548, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.006455600576883456, | |
| "grad_norm": 2.889204740524292, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.6571, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.006592953780646934, | |
| "grad_norm": 2.7503957748413086, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.3354, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.006730306984410411, | |
| "grad_norm": 2.1261167526245117, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.3084, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.006867660188173889, | |
| "grad_norm": 2.0925722122192383, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.3023, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.007005013391937367, | |
| "grad_norm": 2.4151270389556885, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.2791, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.007142366595700844, | |
| "grad_norm": 1.4317853450775146, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.1694, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.0072797197994643225, | |
| "grad_norm": 2.1933341026306152, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.1593, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0074170730032278006, | |
| "grad_norm": 2.4327194690704346, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.1166, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0074170730032278006, | |
| "eval_loss": 0.24664896726608276, | |
| "eval_runtime": 1386.7037, | |
| "eval_samples_per_second": 4.421, | |
| "eval_steps_per_second": 0.553, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.007554426206991278, | |
| "grad_norm": 5.130263805389404, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3211, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.007691779410754756, | |
| "grad_norm": 3.5147716999053955, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.1736, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.007829132614518234, | |
| "grad_norm": 3.190122604370117, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.1557, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.007966485818281712, | |
| "grad_norm": 1.4721782207489014, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0683, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.008103839022045188, | |
| "grad_norm": 3.573559284210205, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.1372, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.008241192225808667, | |
| "grad_norm": 5.100113868713379, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.1927, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.008378545429572145, | |
| "grad_norm": 4.845905303955078, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.128, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.008515898633335623, | |
| "grad_norm": 4.2077741622924805, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.1291, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.0086532518370991, | |
| "grad_norm": 4.020180702209473, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.1386, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0086532518370991, | |
| "eval_loss": 0.16387590765953064, | |
| "eval_runtime": 1386.6269, | |
| "eval_samples_per_second": 4.422, | |
| "eval_steps_per_second": 0.553, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.008790605040862579, | |
| "grad_norm": 4.362114429473877, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.179, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.008927958244626055, | |
| "grad_norm": 3.658870220184326, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.2027, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.009065311448389533, | |
| "grad_norm": 3.2760026454925537, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.1848, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.009202664652153011, | |
| "grad_norm": 2.835029363632202, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.1303, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.00934001785591649, | |
| "grad_norm": 6.923901557922363, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.4301, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.009477371059679968, | |
| "grad_norm": 2.7901358604431152, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.075, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.009614724263443444, | |
| "grad_norm": 4.898245334625244, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.2811, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.009752077467206922, | |
| "grad_norm": 0.8654341101646423, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.0187, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0098894306709704, | |
| "grad_norm": 5.90964937210083, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0876, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0098894306709704, | |
| "eval_loss": 0.12309632450342178, | |
| "eval_runtime": 1386.7026, | |
| "eval_samples_per_second": 4.421, | |
| "eval_steps_per_second": 0.553, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.010026783874733878, | |
| "grad_norm": 2.174041986465454, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0703, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.010164137078497356, | |
| "grad_norm": 0.2941243052482605, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0076, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.010301490282260834, | |
| "grad_norm": 2.9184627532958984, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0765, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.918303489687552e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |