| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.22123893805309736, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0029498525073746312, | |
| "grad_norm": 2.6695961952209473, | |
| "learning_rate": 1e-05, | |
| "loss": 4.059, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0029498525073746312, | |
| "eval_loss": 2.0677247047424316, | |
| "eval_runtime": 44.3994, | |
| "eval_samples_per_second": 6.442, | |
| "eval_steps_per_second": 0.811, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0058997050147492625, | |
| "grad_norm": 1.780763030052185, | |
| "learning_rate": 2e-05, | |
| "loss": 3.1111, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008849557522123894, | |
| "grad_norm": 3.134167194366455, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5404, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.011799410029498525, | |
| "grad_norm": 4.528249263763428, | |
| "learning_rate": 4e-05, | |
| "loss": 4.5936, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.014749262536873156, | |
| "grad_norm": 3.391472339630127, | |
| "learning_rate": 5e-05, | |
| "loss": 4.1795, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.017699115044247787, | |
| "grad_norm": 2.1838064193725586, | |
| "learning_rate": 6e-05, | |
| "loss": 3.0666, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02064896755162242, | |
| "grad_norm": 2.6701786518096924, | |
| "learning_rate": 7e-05, | |
| "loss": 3.5408, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02359882005899705, | |
| "grad_norm": 2.9336118698120117, | |
| "learning_rate": 8e-05, | |
| "loss": 3.67, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02654867256637168, | |
| "grad_norm": 5.5627264976501465, | |
| "learning_rate": 9e-05, | |
| "loss": 3.7612, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02654867256637168, | |
| "eval_loss": 1.4617587327957153, | |
| "eval_runtime": 44.4212, | |
| "eval_samples_per_second": 6.438, | |
| "eval_steps_per_second": 0.81, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.029498525073746312, | |
| "grad_norm": 3.2141149044036865, | |
| "learning_rate": 0.0001, | |
| "loss": 3.1932, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.032448377581120944, | |
| "grad_norm": 2.183711051940918, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.4995, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.035398230088495575, | |
| "grad_norm": 2.1251938343048096, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.4882, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.038348082595870206, | |
| "grad_norm": 2.368736743927002, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 1.8657, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04129793510324484, | |
| "grad_norm": 3.1230757236480713, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.2617, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04424778761061947, | |
| "grad_norm": 2.4695842266082764, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.24, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0471976401179941, | |
| "grad_norm": 3.388068199157715, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.9291, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05014749262536873, | |
| "grad_norm": 3.315546989440918, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.5513, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.05309734513274336, | |
| "grad_norm": 3.138029098510742, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.061, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05309734513274336, | |
| "eval_loss": 1.0333534479141235, | |
| "eval_runtime": 44.4437, | |
| "eval_samples_per_second": 6.435, | |
| "eval_steps_per_second": 0.81, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05604719764011799, | |
| "grad_norm": 2.192615509033203, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.6975, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.058997050147492625, | |
| "grad_norm": 2.5154130458831787, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.3443, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.061946902654867256, | |
| "grad_norm": 2.76058030128479, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.3646, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.06489675516224189, | |
| "grad_norm": 1.8679944276809692, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.1289, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06784660766961652, | |
| "grad_norm": 1.9377188682556152, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.1156, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.07079646017699115, | |
| "grad_norm": 2.5624561309814453, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.9728, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07374631268436578, | |
| "grad_norm": 1.7926393747329712, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.9014, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.07669616519174041, | |
| "grad_norm": 1.742391586303711, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.5669, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07964601769911504, | |
| "grad_norm": 2.4783685207366943, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.2309, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07964601769911504, | |
| "eval_loss": 1.0076097249984741, | |
| "eval_runtime": 44.4736, | |
| "eval_samples_per_second": 6.431, | |
| "eval_steps_per_second": 0.809, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.08259587020648967, | |
| "grad_norm": 2.483924388885498, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.7079, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0855457227138643, | |
| "grad_norm": 2.8759710788726807, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.1136, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.08849557522123894, | |
| "grad_norm": 2.6359269618988037, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.9826, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09144542772861357, | |
| "grad_norm": 1.697524905204773, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.5094, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0943952802359882, | |
| "grad_norm": 3.194821834564209, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.8343, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09734513274336283, | |
| "grad_norm": 3.587616443634033, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.7127, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.10029498525073746, | |
| "grad_norm": 2.2063448429107666, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.7964, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.10324483775811209, | |
| "grad_norm": 2.4330263137817383, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 1.4703, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.10619469026548672, | |
| "grad_norm": 4.004213333129883, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.8824, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10619469026548672, | |
| "eval_loss": 0.9778890609741211, | |
| "eval_runtime": 44.3941, | |
| "eval_samples_per_second": 6.442, | |
| "eval_steps_per_second": 0.811, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10914454277286136, | |
| "grad_norm": 2.327597141265869, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.1188, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.11209439528023599, | |
| "grad_norm": 2.4239344596862793, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 1.8542, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.11504424778761062, | |
| "grad_norm": 2.819444417953491, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.0174, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.11799410029498525, | |
| "grad_norm": 2.2955985069274902, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.8439, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12094395280235988, | |
| "grad_norm": 2.412621021270752, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 1.3983, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.12389380530973451, | |
| "grad_norm": 2.1134157180786133, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.4839, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.12684365781710916, | |
| "grad_norm": 3.077287197113037, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 1.7456, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.12979351032448377, | |
| "grad_norm": 2.8075761795043945, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.6592, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.13274336283185842, | |
| "grad_norm": 3.0821831226348877, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.0796, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.13274336283185842, | |
| "eval_loss": 0.9554188251495361, | |
| "eval_runtime": 44.4051, | |
| "eval_samples_per_second": 6.441, | |
| "eval_steps_per_second": 0.811, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.13569321533923304, | |
| "grad_norm": 1.895007848739624, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.5245, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.13864306784660768, | |
| "grad_norm": 1.7392289638519287, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 1.772, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1415929203539823, | |
| "grad_norm": 2.4289803504943848, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.5995, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.14454277286135694, | |
| "grad_norm": 2.307551383972168, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.3335, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.14749262536873156, | |
| "grad_norm": 2.0941660404205322, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.5142, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1504424778761062, | |
| "grad_norm": 1.5647395849227905, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.0823, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.15339233038348082, | |
| "grad_norm": 3.6214590072631836, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.4263, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.15634218289085547, | |
| "grad_norm": 2.2273218631744385, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 1.8634, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.1592920353982301, | |
| "grad_norm": 2.075181722640991, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.1057, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1592920353982301, | |
| "eval_loss": 0.9436812996864319, | |
| "eval_runtime": 44.4269, | |
| "eval_samples_per_second": 6.438, | |
| "eval_steps_per_second": 0.81, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.16224188790560473, | |
| "grad_norm": 2.0070507526397705, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7728, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.16519174041297935, | |
| "grad_norm": 2.2338101863861084, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.7216, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.168141592920354, | |
| "grad_norm": 2.3538854122161865, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.6894, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.1710914454277286, | |
| "grad_norm": 2.551027536392212, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 1.4082, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.17404129793510326, | |
| "grad_norm": 2.7152366638183594, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 1.7138, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.17699115044247787, | |
| "grad_norm": 3.2682361602783203, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.0795, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.17994100294985252, | |
| "grad_norm": 1.910207748413086, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 2.4073, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.18289085545722714, | |
| "grad_norm": 2.7701568603515625, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 1.678, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.18584070796460178, | |
| "grad_norm": 2.447645425796509, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.9186, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.18584070796460178, | |
| "eval_loss": 0.9331210255622864, | |
| "eval_runtime": 44.3944, | |
| "eval_samples_per_second": 6.442, | |
| "eval_steps_per_second": 0.811, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.1887905604719764, | |
| "grad_norm": 2.811461925506592, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.0873, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.19174041297935104, | |
| "grad_norm": 2.722278594970703, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 1.8269, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.19469026548672566, | |
| "grad_norm": 2.2737720012664795, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.8935, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.1976401179941003, | |
| "grad_norm": 1.9448575973510742, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 1.4455, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.20058997050147492, | |
| "grad_norm": 2.384688138961792, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 1.9643, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.20353982300884957, | |
| "grad_norm": 2.049166440963745, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.3775, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.20648967551622419, | |
| "grad_norm": 2.747662305831909, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 1.7945, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.20943952802359883, | |
| "grad_norm": 2.058532953262329, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 1.4522, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.21238938053097345, | |
| "grad_norm": 2.542921543121338, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.0277, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.21238938053097345, | |
| "eval_loss": 0.9295855760574341, | |
| "eval_runtime": 44.4249, | |
| "eval_samples_per_second": 6.438, | |
| "eval_steps_per_second": 0.81, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2153392330383481, | |
| "grad_norm": 2.58382248878479, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.7377, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.2182890855457227, | |
| "grad_norm": 2.8606531620025635, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 1.8194, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.22123893805309736, | |
| "grad_norm": 3.316101551055908, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.3525, | |
| "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": 9.521684545536e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |