| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2857142857142857, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0038095238095238095, | |
| "grad_norm": 0.7823566794395447, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9022, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0038095238095238095, | |
| "eval_loss": 3.012644052505493, | |
| "eval_runtime": 23.4717, | |
| "eval_samples_per_second": 37.662, | |
| "eval_steps_per_second": 1.193, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007619047619047619, | |
| "grad_norm": 0.8248893022537231, | |
| "learning_rate": 2e-05, | |
| "loss": 2.8666, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.011428571428571429, | |
| "grad_norm": 0.8289470672607422, | |
| "learning_rate": 3e-05, | |
| "loss": 2.9933, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.015238095238095238, | |
| "grad_norm": 0.7047882080078125, | |
| "learning_rate": 4e-05, | |
| "loss": 2.7877, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01904761904761905, | |
| "grad_norm": 0.6757094264030457, | |
| "learning_rate": 5e-05, | |
| "loss": 2.8138, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.022857142857142857, | |
| "grad_norm": 0.851570188999176, | |
| "learning_rate": 6e-05, | |
| "loss": 3.0114, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02666666666666667, | |
| "grad_norm": 0.841671347618103, | |
| "learning_rate": 7e-05, | |
| "loss": 2.8939, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.030476190476190476, | |
| "grad_norm": 0.829535961151123, | |
| "learning_rate": 8e-05, | |
| "loss": 2.936, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03428571428571429, | |
| "grad_norm": 0.9249017238616943, | |
| "learning_rate": 9e-05, | |
| "loss": 3.061, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03428571428571429, | |
| "eval_loss": 2.759498357772827, | |
| "eval_runtime": 23.5193, | |
| "eval_samples_per_second": 37.586, | |
| "eval_steps_per_second": 1.191, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0380952380952381, | |
| "grad_norm": 0.6803405284881592, | |
| "learning_rate": 0.0001, | |
| "loss": 2.6938, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0419047619047619, | |
| "grad_norm": 0.6932120323181152, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.5926, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.045714285714285714, | |
| "grad_norm": 0.8588330149650574, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.6154, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.049523809523809526, | |
| "grad_norm": 1.0089638233184814, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.4526, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.05333333333333334, | |
| "grad_norm": 1.0133357048034668, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.6542, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05714285714285714, | |
| "grad_norm": 0.9814991354942322, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.5002, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06095238095238095, | |
| "grad_norm": 0.7550845146179199, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.4295, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.06476190476190476, | |
| "grad_norm": 0.7560195326805115, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.438, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06857142857142857, | |
| "grad_norm": 0.7135016918182373, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.4482, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06857142857142857, | |
| "eval_loss": 2.444065809249878, | |
| "eval_runtime": 23.4898, | |
| "eval_samples_per_second": 37.633, | |
| "eval_steps_per_second": 1.192, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07238095238095238, | |
| "grad_norm": 0.6023127436637878, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.2937, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0761904761904762, | |
| "grad_norm": 0.7713736295700073, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.2259, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.7127986550331116, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.39, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0838095238095238, | |
| "grad_norm": 0.6863518357276917, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.3785, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08761904761904762, | |
| "grad_norm": 0.6800816655158997, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.261, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.09142857142857143, | |
| "grad_norm": 0.6998750567436218, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.2527, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 0.7263910174369812, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.4401, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09904761904761905, | |
| "grad_norm": 0.675186276435852, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.2137, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.10285714285714286, | |
| "grad_norm": 0.7505806684494019, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.1759, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10285714285714286, | |
| "eval_loss": 2.3132174015045166, | |
| "eval_runtime": 23.4422, | |
| "eval_samples_per_second": 37.71, | |
| "eval_steps_per_second": 1.194, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10666666666666667, | |
| "grad_norm": 0.6597660183906555, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.2907, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.11047619047619048, | |
| "grad_norm": 0.644054651260376, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.2149, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": 0.6115710735321045, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.1749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1180952380952381, | |
| "grad_norm": 0.6088316440582275, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.3669, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1219047619047619, | |
| "grad_norm": 0.7091630697250366, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 2.2771, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.12571428571428572, | |
| "grad_norm": 0.7167139053344727, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.288, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1295238095238095, | |
| "grad_norm": 0.7069408893585205, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.2378, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.6352157592773438, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.2113, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "grad_norm": 0.6456618905067444, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.3842, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "eval_loss": 2.263944387435913, | |
| "eval_runtime": 23.587, | |
| "eval_samples_per_second": 37.478, | |
| "eval_steps_per_second": 1.187, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14095238095238094, | |
| "grad_norm": 0.5665048360824585, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.1493, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.14476190476190476, | |
| "grad_norm": 0.5799903273582458, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.1964, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.14857142857142858, | |
| "grad_norm": 0.6954905390739441, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.1839, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1523809523809524, | |
| "grad_norm": 0.5585700869560242, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.2101, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15619047619047619, | |
| "grad_norm": 0.5499184727668762, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.2176, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.6976773738861084, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.3226, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.16380952380952382, | |
| "grad_norm": 0.5846168398857117, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.1711, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.1676190476190476, | |
| "grad_norm": 1.6517841815948486, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 2.3884, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "grad_norm": 0.5167840719223022, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.973, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17142857142857143, | |
| "eval_loss": 2.2389347553253174, | |
| "eval_runtime": 23.4106, | |
| "eval_samples_per_second": 37.761, | |
| "eval_steps_per_second": 1.196, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17523809523809525, | |
| "grad_norm": 0.5340649485588074, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.1735, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.17904761904761904, | |
| "grad_norm": 0.7212468385696411, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.3213, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "grad_norm": 0.6065192818641663, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.2482, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.18666666666666668, | |
| "grad_norm": 0.6753005385398865, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.2528, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 0.6185556650161743, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.1279, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19428571428571428, | |
| "grad_norm": 0.532867968082428, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.1911, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.1980952380952381, | |
| "grad_norm": 0.6397295594215393, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.1223, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.2019047619047619, | |
| "grad_norm": 0.6476990580558777, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 2.1393, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.2057142857142857, | |
| "grad_norm": 1.3416658639907837, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.2902, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2057142857142857, | |
| "eval_loss": 2.222794771194458, | |
| "eval_runtime": 23.4201, | |
| "eval_samples_per_second": 37.745, | |
| "eval_steps_per_second": 1.196, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.20952380952380953, | |
| "grad_norm": 0.6062219142913818, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2142, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.21333333333333335, | |
| "grad_norm": 0.531572163105011, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.1585, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.21714285714285714, | |
| "grad_norm": 0.5543264746665955, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.2081, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.22095238095238096, | |
| "grad_norm": 0.9360385537147522, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.0396, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.22476190476190477, | |
| "grad_norm": 0.6253273487091064, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.2449, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": 0.6404938697814941, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.2385, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23238095238095238, | |
| "grad_norm": 0.6263427138328552, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 2.1528, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.2361904761904762, | |
| "grad_norm": 0.7024621963500977, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 2.2712, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.5970602631568909, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.1885, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 2.220242977142334, | |
| "eval_runtime": 23.4283, | |
| "eval_samples_per_second": 37.732, | |
| "eval_steps_per_second": 1.195, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2438095238095238, | |
| "grad_norm": 0.631573498249054, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.2119, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.24761904761904763, | |
| "grad_norm": 0.6913384795188904, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.1587, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.25142857142857145, | |
| "grad_norm": 0.6332992911338806, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.1551, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.25523809523809526, | |
| "grad_norm": 0.6200920939445496, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 2.1621, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.259047619047619, | |
| "grad_norm": 0.623578667640686, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 2.1721, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.26285714285714284, | |
| "grad_norm": 0.5753321051597595, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.298, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 1.7991951704025269, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.1777, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2704761904761905, | |
| "grad_norm": 0.6142107248306274, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 2.236, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.2742857142857143, | |
| "grad_norm": 0.5913622379302979, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.2797, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2742857142857143, | |
| "eval_loss": 2.2107057571411133, | |
| "eval_runtime": 23.4571, | |
| "eval_samples_per_second": 37.686, | |
| "eval_steps_per_second": 1.194, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2780952380952381, | |
| "grad_norm": 0.5493815541267395, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.1707, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.2819047619047619, | |
| "grad_norm": 7.019804000854492, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 2.0899, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.6025744676589966, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.1821, | |
| "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": 4.451323701362688e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |