| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04562043795620438, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006082725060827251, | |
| "grad_norm": 5.882946014404297, | |
| "learning_rate": 1e-05, | |
| "loss": 6.8126, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006082725060827251, | |
| "eval_loss": 3.5069375038146973, | |
| "eval_runtime": 250.2929, | |
| "eval_samples_per_second": 5.534, | |
| "eval_steps_per_second": 0.695, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012165450121654502, | |
| "grad_norm": 6.731106758117676, | |
| "learning_rate": 2e-05, | |
| "loss": 7.4335, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0018248175182481751, | |
| "grad_norm": 5.725454807281494, | |
| "learning_rate": 3e-05, | |
| "loss": 7.6123, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0024330900243309003, | |
| "grad_norm": 6.520483016967773, | |
| "learning_rate": 4e-05, | |
| "loss": 7.0299, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0030413625304136255, | |
| "grad_norm": 6.306111812591553, | |
| "learning_rate": 5e-05, | |
| "loss": 6.8649, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0036496350364963502, | |
| "grad_norm": 6.5682291984558105, | |
| "learning_rate": 6e-05, | |
| "loss": 7.7172, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004257907542579075, | |
| "grad_norm": 8.165228843688965, | |
| "learning_rate": 7e-05, | |
| "loss": 6.3985, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004866180048661801, | |
| "grad_norm": 7.165604591369629, | |
| "learning_rate": 8e-05, | |
| "loss": 6.2562, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005474452554744526, | |
| "grad_norm": 10.563287734985352, | |
| "learning_rate": 9e-05, | |
| "loss": 5.8828, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005474452554744526, | |
| "eval_loss": 2.5655674934387207, | |
| "eval_runtime": 251.4954, | |
| "eval_samples_per_second": 5.507, | |
| "eval_steps_per_second": 0.692, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006082725060827251, | |
| "grad_norm": 6.582736492156982, | |
| "learning_rate": 0.0001, | |
| "loss": 4.7274, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006690997566909975, | |
| "grad_norm": 6.260922908782959, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 4.3927, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0072992700729927005, | |
| "grad_norm": 8.289908409118652, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.5041, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007907542579075427, | |
| "grad_norm": 6.218399524688721, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 4.0505, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00851581508515815, | |
| "grad_norm": 11.364131927490234, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.858, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009124087591240875, | |
| "grad_norm": 7.5929341316223145, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.8467, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009732360097323601, | |
| "grad_norm": 8.452181816101074, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 3.3068, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.010340632603406326, | |
| "grad_norm": 10.253931045532227, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.2907, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.010948905109489052, | |
| "grad_norm": 7.085786819458008, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.6662, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.010948905109489052, | |
| "eval_loss": 1.0290194749832153, | |
| "eval_runtime": 252.348, | |
| "eval_samples_per_second": 5.488, | |
| "eval_steps_per_second": 0.69, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011557177615571776, | |
| "grad_norm": 6.795413017272949, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.5245, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.012165450121654502, | |
| "grad_norm": 4.914115905761719, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.2945, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012773722627737226, | |
| "grad_norm": 7.551743030548096, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.6129, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01338199513381995, | |
| "grad_norm": 7.432616233825684, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 1.3972, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.013990267639902677, | |
| "grad_norm": 7.51466703414917, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.5659, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.014598540145985401, | |
| "grad_norm": 7.66641902923584, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6989, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.015206812652068127, | |
| "grad_norm": 7.0404582023620605, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.5088, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.015815085158150853, | |
| "grad_norm": 5.111483573913574, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.0722, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.016423357664233577, | |
| "grad_norm": 6.778738498687744, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.4215, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016423357664233577, | |
| "eval_loss": 0.6313769221305847, | |
| "eval_runtime": 251.1945, | |
| "eval_samples_per_second": 5.514, | |
| "eval_steps_per_second": 0.693, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0170316301703163, | |
| "grad_norm": 5.726966381072998, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.2793, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.017639902676399026, | |
| "grad_norm": 9.59007453918457, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.0635, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01824817518248175, | |
| "grad_norm": 7.323919296264648, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.2779, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018856447688564478, | |
| "grad_norm": 9.515922546386719, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.2738, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.019464720194647202, | |
| "grad_norm": 6.336705207824707, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.897, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.020072992700729927, | |
| "grad_norm": 4.728940963745117, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.9216, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02068126520681265, | |
| "grad_norm": 5.460188865661621, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.0803, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.021289537712895375, | |
| "grad_norm": 5.335496425628662, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 1.2525, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.021897810218978103, | |
| "grad_norm": 7.545004844665527, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.9629, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.021897810218978103, | |
| "eval_loss": 0.5044350624084473, | |
| "eval_runtime": 251.2533, | |
| "eval_samples_per_second": 5.512, | |
| "eval_steps_per_second": 0.693, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.022506082725060828, | |
| "grad_norm": 8.623957633972168, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.3611, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.023114355231143552, | |
| "grad_norm": 4.0373735427856445, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.5133, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.023722627737226276, | |
| "grad_norm": 5.937623500823975, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.0959, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.024330900243309004, | |
| "grad_norm": 4.628834247589111, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.612, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.024939172749391728, | |
| "grad_norm": 4.886986255645752, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 1.1789, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.025547445255474453, | |
| "grad_norm": 6.197083950042725, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.2511, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.026155717761557177, | |
| "grad_norm": 4.846970558166504, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.7351, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0267639902676399, | |
| "grad_norm": 19.016860961914062, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.2659, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02737226277372263, | |
| "grad_norm": 5.002533435821533, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.974, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02737226277372263, | |
| "eval_loss": 0.501889705657959, | |
| "eval_runtime": 251.1579, | |
| "eval_samples_per_second": 5.514, | |
| "eval_steps_per_second": 0.693, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.027980535279805353, | |
| "grad_norm": 16.9000301361084, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.1102, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.028588807785888078, | |
| "grad_norm": 4.386449813842773, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.7829, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.029197080291970802, | |
| "grad_norm": 3.59611177444458, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.7655, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.029805352798053526, | |
| "grad_norm": 4.8466668128967285, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.9544, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.030413625304136254, | |
| "grad_norm": 4.518596649169922, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.6498, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03102189781021898, | |
| "grad_norm": 5.2425336837768555, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.0189, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.031630170316301706, | |
| "grad_norm": 3.82919979095459, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.5889, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.03223844282238443, | |
| "grad_norm": 3.4996325969696045, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.57, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.032846715328467155, | |
| "grad_norm": 4.0521955490112305, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.8676, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.032846715328467155, | |
| "eval_loss": 0.4546598196029663, | |
| "eval_runtime": 251.2833, | |
| "eval_samples_per_second": 5.512, | |
| "eval_steps_per_second": 0.692, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03345498783454988, | |
| "grad_norm": 4.143316268920898, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8453, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0340632603406326, | |
| "grad_norm": 3.8293848037719727, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.6753, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.03467153284671533, | |
| "grad_norm": 4.145463466644287, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.8705, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03527980535279805, | |
| "grad_norm": 2.8706634044647217, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.5544, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.035888077858880776, | |
| "grad_norm": 4.694822788238525, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 1.203, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0364963503649635, | |
| "grad_norm": 4.350282192230225, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.9564, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03710462287104623, | |
| "grad_norm": 3.686616897583008, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.7235, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.037712895377128956, | |
| "grad_norm": 3.8983712196350098, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.9513, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.03832116788321168, | |
| "grad_norm": 3.7569422721862793, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.7914, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03832116788321168, | |
| "eval_loss": 0.4329450726509094, | |
| "eval_runtime": 251.5697, | |
| "eval_samples_per_second": 5.505, | |
| "eval_steps_per_second": 0.692, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.038929440389294405, | |
| "grad_norm": 7.999822616577148, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.3853, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.03953771289537713, | |
| "grad_norm": 4.116574764251709, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 1.1761, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.040145985401459854, | |
| "grad_norm": 2.7404329776763916, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.8501, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.04075425790754258, | |
| "grad_norm": 4.554178237915039, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.7249, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0413625304136253, | |
| "grad_norm": 4.6993536949157715, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 1.1265, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.041970802919708027, | |
| "grad_norm": 5.122533798217773, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.3654, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.04257907542579075, | |
| "grad_norm": 4.0280842781066895, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.9958, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.04318734793187348, | |
| "grad_norm": 3.3179454803466797, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.7935, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.043795620437956206, | |
| "grad_norm": 3.2039504051208496, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.5538, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.043795620437956206, | |
| "eval_loss": 0.42027562856674194, | |
| "eval_runtime": 251.1143, | |
| "eval_samples_per_second": 5.515, | |
| "eval_steps_per_second": 0.693, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04440389294403893, | |
| "grad_norm": 2.3764936923980713, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.5485, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.045012165450121655, | |
| "grad_norm": 4.495674133300781, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.9866, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.04562043795620438, | |
| "grad_norm": 4.754255294799805, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.8529, | |
| "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": 9.80187302854656e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |