| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12953367875647667, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025906735751295338, | |
| "grad_norm": 3.7401230335235596, | |
| "learning_rate": 1e-05, | |
| "loss": 3.0666, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0025906735751295338, | |
| "eval_loss": 4.18491792678833, | |
| "eval_runtime": 35.7075, | |
| "eval_samples_per_second": 9.102, | |
| "eval_steps_per_second": 1.148, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0051813471502590676, | |
| "grad_norm": 4.653674125671387, | |
| "learning_rate": 2e-05, | |
| "loss": 4.1643, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007772020725388601, | |
| "grad_norm": 5.061185836791992, | |
| "learning_rate": 3e-05, | |
| "loss": 3.7233, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010362694300518135, | |
| "grad_norm": 4.979389667510986, | |
| "learning_rate": 4e-05, | |
| "loss": 4.6052, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.012953367875647668, | |
| "grad_norm": 4.482627868652344, | |
| "learning_rate": 5e-05, | |
| "loss": 3.4867, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015544041450777202, | |
| "grad_norm": 6.006535053253174, | |
| "learning_rate": 6e-05, | |
| "loss": 3.7069, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018134715025906734, | |
| "grad_norm": 4.89888334274292, | |
| "learning_rate": 7e-05, | |
| "loss": 3.9644, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02072538860103627, | |
| "grad_norm": 5.465033531188965, | |
| "learning_rate": 8e-05, | |
| "loss": 3.3032, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.023316062176165803, | |
| "grad_norm": 4.801088809967041, | |
| "learning_rate": 9e-05, | |
| "loss": 2.0563, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.023316062176165803, | |
| "eval_loss": 3.25215744972229, | |
| "eval_runtime": 35.2692, | |
| "eval_samples_per_second": 9.215, | |
| "eval_steps_per_second": 1.162, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025906735751295335, | |
| "grad_norm": 8.266736030578613, | |
| "learning_rate": 0.0001, | |
| "loss": 3.6629, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02849740932642487, | |
| "grad_norm": 4.273421287536621, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 1.7825, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.031088082901554404, | |
| "grad_norm": 9.370466232299805, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.9157, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03367875647668394, | |
| "grad_norm": 5.781856536865234, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.4527, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03626943005181347, | |
| "grad_norm": 7.081896781921387, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.0791, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.038860103626943004, | |
| "grad_norm": 9.198928833007812, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.8983, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04145077720207254, | |
| "grad_norm": 7.364632606506348, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.0986, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04404145077720207, | |
| "grad_norm": 5.105291366577148, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 1.4194, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.046632124352331605, | |
| "grad_norm": 7.902851104736328, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.5238, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.046632124352331605, | |
| "eval_loss": 2.1568806171417236, | |
| "eval_runtime": 35.3004, | |
| "eval_samples_per_second": 9.207, | |
| "eval_steps_per_second": 1.161, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04922279792746114, | |
| "grad_norm": 7.482724666595459, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.0681, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05181347150259067, | |
| "grad_norm": 6.279477119445801, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.0164, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.054404145077720206, | |
| "grad_norm": 6.212917804718018, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.9994, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05699481865284974, | |
| "grad_norm": 5.9393815994262695, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.2696, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.05958549222797927, | |
| "grad_norm": 6.2451171875, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.0264, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06217616580310881, | |
| "grad_norm": 4.755683898925781, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.7309, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06476683937823834, | |
| "grad_norm": 5.793985366821289, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.432, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06735751295336788, | |
| "grad_norm": 6.06044864654541, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.6466, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06994818652849741, | |
| "grad_norm": 5.624882698059082, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.5609, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06994818652849741, | |
| "eval_loss": 1.9685734510421753, | |
| "eval_runtime": 35.2965, | |
| "eval_samples_per_second": 9.208, | |
| "eval_steps_per_second": 1.162, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07253886010362694, | |
| "grad_norm": 5.412827968597412, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.9472, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07512953367875648, | |
| "grad_norm": 8.074312210083008, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.8224, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07772020725388601, | |
| "grad_norm": 10.367136001586914, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.7846, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08031088082901554, | |
| "grad_norm": 5.971827507019043, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 1.8626, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08290155440414508, | |
| "grad_norm": 6.0518975257873535, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.9026, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08549222797927461, | |
| "grad_norm": 7.365506172180176, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.6508, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08808290155440414, | |
| "grad_norm": 7.813849925994873, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.2382, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09067357512953368, | |
| "grad_norm": 6.097650527954102, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 1.8182, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09326424870466321, | |
| "grad_norm": 8.286858558654785, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.0077, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09326424870466321, | |
| "eval_loss": 1.8582597970962524, | |
| "eval_runtime": 35.2881, | |
| "eval_samples_per_second": 9.21, | |
| "eval_steps_per_second": 1.162, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09585492227979274, | |
| "grad_norm": 7.306047439575195, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.754, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09844559585492228, | |
| "grad_norm": 8.38849925994873, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 1.59, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10103626943005181, | |
| "grad_norm": 5.015171527862549, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.272, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10362694300518134, | |
| "grad_norm": 7.593957901000977, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.1336, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10621761658031088, | |
| "grad_norm": 9.562592506408691, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.4349, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10880829015544041, | |
| "grad_norm": 7.189981937408447, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.827, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11139896373056994, | |
| "grad_norm": 5.631102561950684, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 1.5876, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11398963730569948, | |
| "grad_norm": 5.808564186096191, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.7378, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11658031088082901, | |
| "grad_norm": 6.575247764587402, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.4891, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11658031088082901, | |
| "eval_loss": 1.8105183839797974, | |
| "eval_runtime": 35.2968, | |
| "eval_samples_per_second": 9.208, | |
| "eval_steps_per_second": 1.162, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11917098445595854, | |
| "grad_norm": 6.088664531707764, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.4166, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12176165803108809, | |
| "grad_norm": 5.614904403686523, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 1.6618, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12435233160621761, | |
| "grad_norm": 5.483867645263672, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.0089, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12694300518134716, | |
| "grad_norm": 5.806712627410889, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 1.3856, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.12953367875647667, | |
| "grad_norm": 6.254551410675049, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.0932, | |
| "step": 50 | |
| } | |
| ], | |
| "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": 3.70943641780224e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |