| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.030916679548616478, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006183335909723295, | |
| "grad_norm": 0.8156535625457764, | |
| "learning_rate": 5e-06, | |
| "loss": 7.2519, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006183335909723295, | |
| "eval_loss": 1.9691991806030273, | |
| "eval_runtime": 36.6847, | |
| "eval_samples_per_second": 18.564, | |
| "eval_steps_per_second": 9.295, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001236667181944659, | |
| "grad_norm": 1.1488978862762451, | |
| "learning_rate": 1e-05, | |
| "loss": 6.9605, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0018550007729169888, | |
| "grad_norm": 1.3825483322143555, | |
| "learning_rate": 1.5e-05, | |
| "loss": 6.988, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002473334363889318, | |
| "grad_norm": 1.8649797439575195, | |
| "learning_rate": 2e-05, | |
| "loss": 8.0611, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003091667954861648, | |
| "grad_norm": 0.9418125748634338, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.141, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0037100015458339775, | |
| "grad_norm": 1.8627277612686157, | |
| "learning_rate": 3e-05, | |
| "loss": 8.4816, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004328335136806307, | |
| "grad_norm": 2.2024805545806885, | |
| "learning_rate": 3.5e-05, | |
| "loss": 8.724, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004946668727778636, | |
| "grad_norm": 1.331282138824463, | |
| "learning_rate": 4e-05, | |
| "loss": 7.7042, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005565002318750966, | |
| "grad_norm": 1.6964207887649536, | |
| "learning_rate": 4.5e-05, | |
| "loss": 8.5007, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006183335909723296, | |
| "grad_norm": 1.71376371383667, | |
| "learning_rate": 5e-05, | |
| "loss": 7.5547, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006801669500695625, | |
| "grad_norm": 1.081741452217102, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 8.1592, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007420003091667955, | |
| "grad_norm": 2.098921537399292, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 8.2364, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008038336682640285, | |
| "grad_norm": 1.5198915004730225, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 6.9091, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008038336682640285, | |
| "eval_loss": 1.9464514255523682, | |
| "eval_runtime": 36.6926, | |
| "eval_samples_per_second": 18.56, | |
| "eval_steps_per_second": 9.293, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.008656670273612613, | |
| "grad_norm": 1.6844040155410767, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 8.6812, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009275003864584944, | |
| "grad_norm": 2.3919613361358643, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 7.9752, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009893337455557273, | |
| "grad_norm": 3.004586935043335, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 8.902, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.010511671046529603, | |
| "grad_norm": 1.3965468406677246, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 6.9089, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011130004637501932, | |
| "grad_norm": 2.1235411167144775, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 7.2692, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011748338228474263, | |
| "grad_norm": 1.202420711517334, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 6.8648, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.012366671819446591, | |
| "grad_norm": 2.219177484512329, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 6.788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01298500541041892, | |
| "grad_norm": 2.773238182067871, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 6.2868, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01360333900139125, | |
| "grad_norm": 1.1058857440948486, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 6.9249, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01422167259236358, | |
| "grad_norm": 1.4601209163665771, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 6.5225, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.01484000618333591, | |
| "grad_norm": 4.9399309158325195, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 7.1664, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.015458339774308239, | |
| "grad_norm": 3.4041075706481934, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 7.4847, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01607667336528057, | |
| "grad_norm": 1.3729993104934692, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 7.1938, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01607667336528057, | |
| "eval_loss": 1.7993429899215698, | |
| "eval_runtime": 36.655, | |
| "eval_samples_per_second": 18.579, | |
| "eval_steps_per_second": 9.303, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.016695006956252898, | |
| "grad_norm": 2.970980405807495, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 6.8607, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.017313340547225227, | |
| "grad_norm": 2.7162935733795166, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 5.9147, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01793167413819756, | |
| "grad_norm": 2.087170362472534, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 7.3246, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.018550007729169888, | |
| "grad_norm": 3.4586870670318604, | |
| "learning_rate": 2.5e-05, | |
| "loss": 6.6356, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.019168341320142217, | |
| "grad_norm": 3.089322090148926, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 5.4756, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.019786674911114546, | |
| "grad_norm": 3.5235681533813477, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 6.9095, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.020405008502086874, | |
| "grad_norm": 2.399798631668091, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 7.0229, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.021023342093059207, | |
| "grad_norm": 1.6243059635162354, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 6.9849, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.021641675684031535, | |
| "grad_norm": 3.6505990028381348, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 6.9756, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.022260009275003864, | |
| "grad_norm": 2.003025770187378, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 6.3971, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.022878342865976193, | |
| "grad_norm": 3.3420064449310303, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 7.2006, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.023496676456948525, | |
| "grad_norm": 3.184504508972168, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 5.2291, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.024115010047920854, | |
| "grad_norm": 1.609082818031311, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 6.2046, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.024115010047920854, | |
| "eval_loss": 1.720837950706482, | |
| "eval_runtime": 36.6645, | |
| "eval_samples_per_second": 18.574, | |
| "eval_steps_per_second": 9.301, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.024733343638893183, | |
| "grad_norm": 2.634270668029785, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 6.2318, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02535167722986551, | |
| "grad_norm": 3.3636908531188965, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 7.7077, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.02597001082083784, | |
| "grad_norm": 2.037014961242676, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 6.2751, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.026588344411810173, | |
| "grad_norm": 3.388461112976074, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 7.6244, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0272066780027825, | |
| "grad_norm": 1.931207299232483, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 5.9563, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02782501159375483, | |
| "grad_norm": 3.204418182373047, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 7.8852, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02844334518472716, | |
| "grad_norm": 3.5687596797943115, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 6.9575, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.02906167877569949, | |
| "grad_norm": 3.0171327590942383, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 7.6163, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.02968001236667182, | |
| "grad_norm": 1.911759614944458, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 6.467, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03029834595764415, | |
| "grad_norm": 2.2276806831359863, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 6.9883, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.030916679548616478, | |
| "grad_norm": 2.965574264526367, | |
| "learning_rate": 0.0, | |
| "loss": 6.6589, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.614774395731968e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |