| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02497736426363608, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002497736426363608, | |
| "eval_loss": 1.0045465230941772, | |
| "eval_runtime": 1280.7655, | |
| "eval_samples_per_second": 10.53, | |
| "eval_steps_per_second": 5.265, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007493209279090824, | |
| "grad_norm": 16.586885452270508, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 30.7389, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0014986418558181648, | |
| "grad_norm": 21.692514419555664, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 32.9669, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002247962783727247, | |
| "grad_norm": 11.184452056884766, | |
| "learning_rate": 6.75e-05, | |
| "loss": 29.6533, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002247962783727247, | |
| "eval_loss": 0.8681882619857788, | |
| "eval_runtime": 1280.9699, | |
| "eval_samples_per_second": 10.528, | |
| "eval_steps_per_second": 5.264, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0029972837116363296, | |
| "grad_norm": 13.137224197387695, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 26.0245, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003746604639545412, | |
| "grad_norm": 19.899791717529297, | |
| "learning_rate": 0.0001125, | |
| "loss": 25.7329, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004495925567454494, | |
| "grad_norm": 10.93524169921875, | |
| "learning_rate": 0.000135, | |
| "loss": 23.219, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004495925567454494, | |
| "eval_loss": 0.7281808853149414, | |
| "eval_runtime": 1283.1233, | |
| "eval_samples_per_second": 10.51, | |
| "eval_steps_per_second": 5.255, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005245246495363577, | |
| "grad_norm": 7.8446173667907715, | |
| "learning_rate": 0.00014994217771805422, | |
| "loss": 23.3888, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005994567423272659, | |
| "grad_norm": 10.064414024353027, | |
| "learning_rate": 0.00014907662554463532, | |
| "loss": 21.0303, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006743888351181742, | |
| "grad_norm": 9.759185791015625, | |
| "learning_rate": 0.0001471841427340235, | |
| "loss": 21.3012, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.006743888351181742, | |
| "eval_loss": 0.6666631698608398, | |
| "eval_runtime": 1282.5291, | |
| "eval_samples_per_second": 10.515, | |
| "eval_steps_per_second": 5.258, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007493209279090824, | |
| "grad_norm": 7.994441509246826, | |
| "learning_rate": 0.0001442909649383465, | |
| "loss": 21.0171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008242530206999906, | |
| "grad_norm": 6.8677191734313965, | |
| "learning_rate": 0.0001404372005304598, | |
| "loss": 19.7584, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.008991851134908989, | |
| "grad_norm": 10.436850547790527, | |
| "learning_rate": 0.00013567627457812106, | |
| "loss": 20.2279, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.008991851134908989, | |
| "eval_loss": 0.6394243836402893, | |
| "eval_runtime": 1283.1086, | |
| "eval_samples_per_second": 10.51, | |
| "eval_steps_per_second": 5.255, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009741172062818071, | |
| "grad_norm": 8.883223533630371, | |
| "learning_rate": 0.0001300741882076764, | |
| "loss": 19.3902, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.010490492990727154, | |
| "grad_norm": 7.911937236785889, | |
| "learning_rate": 0.00012370860362476374, | |
| "loss": 20.3179, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.011239813918636236, | |
| "grad_norm": 8.104355812072754, | |
| "learning_rate": 0.00011666776747647015, | |
| "loss": 21.057, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.011239813918636236, | |
| "eval_loss": 0.6243658065795898, | |
| "eval_runtime": 1281.3393, | |
| "eval_samples_per_second": 10.525, | |
| "eval_steps_per_second": 5.262, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.011989134846545318, | |
| "grad_norm": 8.9419584274292, | |
| "learning_rate": 0.00010904928748046599, | |
| "loss": 19.5364, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.012738455774454401, | |
| "grad_norm": 7.041481971740723, | |
| "learning_rate": 0.00010095877928081196, | |
| "loss": 19.3287, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.013487776702363483, | |
| "grad_norm": 7.952820777893066, | |
| "learning_rate": 9.25084022891929e-05, | |
| "loss": 18.6364, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.013487776702363483, | |
| "eval_loss": 0.6130433678627014, | |
| "eval_runtime": 1280.8668, | |
| "eval_samples_per_second": 10.529, | |
| "eval_steps_per_second": 5.264, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.014237097630272566, | |
| "grad_norm": 8.097841262817383, | |
| "learning_rate": 8.381530480933783e-05, | |
| "loss": 19.0937, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.014986418558181648, | |
| "grad_norm": 7.304646968841553, | |
| "learning_rate": 7.5e-05, | |
| "loss": 20.2986, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01573573948609073, | |
| "grad_norm": 6.223910808563232, | |
| "learning_rate": 6.618469519066217e-05, | |
| "loss": 18.0383, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01573573948609073, | |
| "eval_loss": 0.6045629978179932, | |
| "eval_runtime": 1280.7351, | |
| "eval_samples_per_second": 10.53, | |
| "eval_steps_per_second": 5.265, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.016485060413999813, | |
| "grad_norm": 7.179214000701904, | |
| "learning_rate": 5.7491597710807114e-05, | |
| "loss": 20.7756, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.017234381341908894, | |
| "grad_norm": 7.109331130981445, | |
| "learning_rate": 4.904122071918801e-05, | |
| "loss": 17.879, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.017983702269817978, | |
| "grad_norm": 7.5771565437316895, | |
| "learning_rate": 4.095071251953399e-05, | |
| "loss": 19.8381, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.017983702269817978, | |
| "eval_loss": 0.5989252328872681, | |
| "eval_runtime": 1282.9019, | |
| "eval_samples_per_second": 10.512, | |
| "eval_steps_per_second": 5.256, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.01873302319772706, | |
| "grad_norm": 7.1285858154296875, | |
| "learning_rate": 3.333223252352985e-05, | |
| "loss": 17.8126, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.019482344125636143, | |
| "grad_norm": 7.0520853996276855, | |
| "learning_rate": 2.6291396375236232e-05, | |
| "loss": 18.5479, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.020231665053545223, | |
| "grad_norm": 7.003568172454834, | |
| "learning_rate": 1.99258117923236e-05, | |
| "loss": 18.3124, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.020231665053545223, | |
| "eval_loss": 0.5950871706008911, | |
| "eval_runtime": 1283.325, | |
| "eval_samples_per_second": 10.509, | |
| "eval_steps_per_second": 5.254, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.020980985981454307, | |
| "grad_norm": 6.522275924682617, | |
| "learning_rate": 1.4323725421878949e-05, | |
| "loss": 17.3887, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.021730306909363388, | |
| "grad_norm": 7.3920392990112305, | |
| "learning_rate": 9.56279946954021e-06, | |
| "loss": 18.5704, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.022479627837272472, | |
| "grad_norm": 6.510364055633545, | |
| "learning_rate": 5.709035061653494e-06, | |
| "loss": 18.3766, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.022479627837272472, | |
| "eval_loss": 0.5932875275611877, | |
| "eval_runtime": 1283.0357, | |
| "eval_samples_per_second": 10.511, | |
| "eval_steps_per_second": 5.256, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.023228948765181553, | |
| "grad_norm": 6.111503601074219, | |
| "learning_rate": 2.815857265976462e-06, | |
| "loss": 19.1132, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.023978269693090637, | |
| "grad_norm": 7.059328079223633, | |
| "learning_rate": 9.233744553646754e-07, | |
| "loss": 19.0661, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.024727590620999718, | |
| "grad_norm": 7.499072074890137, | |
| "learning_rate": 5.7822281945782424e-08, | |
| "loss": 18.9632, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.024727590620999718, | |
| "eval_loss": 0.592795193195343, | |
| "eval_runtime": 1284.0687, | |
| "eval_samples_per_second": 10.503, | |
| "eval_steps_per_second": 5.251, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.259104775831552e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |