| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0033729372380703424, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.372937238070342e-05, | |
| "eval_loss": 0.9200180768966675, | |
| "eval_runtime": 1651.9865, | |
| "eval_samples_per_second": 30.226, | |
| "eval_steps_per_second": 3.778, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010118811714211028, | |
| "grad_norm": 0.370296448469162, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9664, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00020237623428422057, | |
| "grad_norm": 0.3390261232852936, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9103, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00030356435142633083, | |
| "grad_norm": 0.2934589087963104, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8309, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00030356435142633083, | |
| "eval_loss": 0.9088784456253052, | |
| "eval_runtime": 1657.5975, | |
| "eval_samples_per_second": 30.124, | |
| "eval_steps_per_second": 3.766, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00040475246856844113, | |
| "grad_norm": 0.3271854519844055, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.8315, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0005059405857105514, | |
| "grad_norm": 0.27277594804763794, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.8816, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006071287028526617, | |
| "grad_norm": 0.29398417472839355, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.8761, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0006071287028526617, | |
| "eval_loss": 0.8483078479766846, | |
| "eval_runtime": 1658.0801, | |
| "eval_samples_per_second": 30.115, | |
| "eval_steps_per_second": 3.765, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0007083168199947719, | |
| "grad_norm": 0.2874149978160858, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.8283, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0008095049371368823, | |
| "grad_norm": 0.2777639329433441, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.7365, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0009106930542789925, | |
| "grad_norm": 0.2506990432739258, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.8006, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0009106930542789925, | |
| "eval_loss": 0.830035924911499, | |
| "eval_runtime": 1657.6879, | |
| "eval_samples_per_second": 30.122, | |
| "eval_steps_per_second": 3.765, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0010118811714211029, | |
| "grad_norm": 0.26121965050697327, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.7908, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001113069288563213, | |
| "grad_norm": 0.22511965036392212, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6791, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0012142574057053233, | |
| "grad_norm": 0.2881379723548889, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.8729, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0012142574057053233, | |
| "eval_loss": 0.8234006762504578, | |
| "eval_runtime": 1657.9322, | |
| "eval_samples_per_second": 30.118, | |
| "eval_steps_per_second": 3.765, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0013154455228474337, | |
| "grad_norm": 0.2760739028453827, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.7797, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0014166336399895438, | |
| "grad_norm": 0.2843494117259979, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.7533, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0015178217571316542, | |
| "grad_norm": 0.23762422800064087, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.7663, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0015178217571316542, | |
| "eval_loss": 0.8193370699882507, | |
| "eval_runtime": 1657.7176, | |
| "eval_samples_per_second": 30.122, | |
| "eval_steps_per_second": 3.765, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0016190098742737645, | |
| "grad_norm": 0.3037606179714203, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.7856, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0017201979914158747, | |
| "grad_norm": 0.2485446184873581, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.7938, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.001821386108557985, | |
| "grad_norm": 0.24273481965065002, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.7606, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.001821386108557985, | |
| "eval_loss": 0.8170143365859985, | |
| "eval_runtime": 1656.2446, | |
| "eval_samples_per_second": 30.148, | |
| "eval_steps_per_second": 3.769, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0019225742257000954, | |
| "grad_norm": 0.2631082534790039, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.7554, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0020237623428422057, | |
| "grad_norm": 0.2768804430961609, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.8178, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.002124950459984316, | |
| "grad_norm": 0.23129825294017792, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.8145, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.002124950459984316, | |
| "eval_loss": 0.8158968091011047, | |
| "eval_runtime": 1657.0323, | |
| "eval_samples_per_second": 30.134, | |
| "eval_steps_per_second": 3.767, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.002226138577126426, | |
| "grad_norm": 0.28922489285469055, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.7805, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0023273266942685365, | |
| "grad_norm": 0.30259978771209717, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.8183, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0024285148114106467, | |
| "grad_norm": 0.2592143714427948, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.8365, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0024285148114106467, | |
| "eval_loss": 0.8151922225952148, | |
| "eval_runtime": 1656.7975, | |
| "eval_samples_per_second": 30.138, | |
| "eval_steps_per_second": 3.768, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.002529702928552757, | |
| "grad_norm": 0.21596688032150269, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.785, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0026308910456948674, | |
| "grad_norm": 0.2620626389980316, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.8722, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0027320791628369775, | |
| "grad_norm": 0.29990634322166443, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.8015, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0027320791628369775, | |
| "eval_loss": 0.8147459030151367, | |
| "eval_runtime": 1656.7133, | |
| "eval_samples_per_second": 30.14, | |
| "eval_steps_per_second": 3.768, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0028332672799790876, | |
| "grad_norm": 0.2881883382797241, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.7495, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.002934455397121198, | |
| "grad_norm": 0.2536965012550354, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.8133, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0030356435142633083, | |
| "grad_norm": 0.28757724165916443, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.7783, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0030356435142633083, | |
| "eval_loss": 0.8145592212677002, | |
| "eval_runtime": 1657.575, | |
| "eval_samples_per_second": 30.124, | |
| "eval_steps_per_second": 3.766, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0031368316314054185, | |
| "grad_norm": 0.24967505037784576, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.7542, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.003238019748547529, | |
| "grad_norm": 0.25066524744033813, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.8054, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.003339207865689639, | |
| "grad_norm": 0.2577332854270935, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.7455, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.003339207865689639, | |
| "eval_loss": 0.8145051598548889, | |
| "eval_runtime": 1658.0359, | |
| "eval_samples_per_second": 30.116, | |
| "eval_steps_per_second": 3.765, | |
| "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": 3.0827944549023744e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |