| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.009493745994825909, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.493745994825909e-05, | |
| "eval_loss": 0.9916548728942871, | |
| "eval_runtime": 301.416, | |
| "eval_samples_per_second": 58.856, | |
| "eval_steps_per_second": 7.359, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002848123798447773, | |
| "grad_norm": 0.11237525939941406, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9762, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005696247596895546, | |
| "grad_norm": 0.11895499378442764, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9471, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008544371395343318, | |
| "grad_norm": 0.115925632417202, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.0212, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008544371395343318, | |
| "eval_loss": 0.9887385964393616, | |
| "eval_runtime": 301.4338, | |
| "eval_samples_per_second": 58.852, | |
| "eval_steps_per_second": 7.358, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011392495193791091, | |
| "grad_norm": 0.12413384765386581, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.9034, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0014240618992238862, | |
| "grad_norm": 0.11049390584230423, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.9668, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0017088742790686636, | |
| "grad_norm": 0.1114693433046341, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.0035, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0017088742790686636, | |
| "eval_loss": 0.965744137763977, | |
| "eval_runtime": 301.4187, | |
| "eval_samples_per_second": 58.855, | |
| "eval_steps_per_second": 7.359, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001993686658913441, | |
| "grad_norm": 0.10345105826854706, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.9413, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0022784990387582182, | |
| "grad_norm": 0.10668003559112549, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.9965, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002563311418602995, | |
| "grad_norm": 0.08716581016778946, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.9437, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002563311418602995, | |
| "eval_loss": 0.9435688853263855, | |
| "eval_runtime": 301.4814, | |
| "eval_samples_per_second": 58.843, | |
| "eval_steps_per_second": 7.357, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0028481237984477725, | |
| "grad_norm": 0.08130133897066116, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.9626, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00313293617829255, | |
| "grad_norm": 0.12710551917552948, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.8768, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003417748558137327, | |
| "grad_norm": 0.1261487901210785, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.9892, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003417748558137327, | |
| "eval_loss": 0.9150912165641785, | |
| "eval_runtime": 301.3752, | |
| "eval_samples_per_second": 58.864, | |
| "eval_steps_per_second": 7.36, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0037025609379821045, | |
| "grad_norm": 0.11289028078317642, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.9106, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003987373317826882, | |
| "grad_norm": 0.1026608943939209, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.8848, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004272185697671659, | |
| "grad_norm": 0.09589796513319016, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.9481, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004272185697671659, | |
| "eval_loss": 0.8930308222770691, | |
| "eval_runtime": 301.9003, | |
| "eval_samples_per_second": 58.761, | |
| "eval_steps_per_second": 7.347, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0045569980775164365, | |
| "grad_norm": 0.07461383938789368, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.9019, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004841810457361213, | |
| "grad_norm": 0.0672895759344101, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.7937, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.00512662283720599, | |
| "grad_norm": 0.0820642039179802, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.9105, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.00512662283720599, | |
| "eval_loss": 0.8874810338020325, | |
| "eval_runtime": 301.5568, | |
| "eval_samples_per_second": 58.828, | |
| "eval_steps_per_second": 7.355, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005411435217050768, | |
| "grad_norm": 0.07211911678314209, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.815, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005696247596895545, | |
| "grad_norm": 0.06929625570774078, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.8154, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005981059976740323, | |
| "grad_norm": 0.06639785319566727, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.8249, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005981059976740323, | |
| "eval_loss": 0.8842635750770569, | |
| "eval_runtime": 300.8883, | |
| "eval_samples_per_second": 58.959, | |
| "eval_steps_per_second": 7.372, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0062658723565851, | |
| "grad_norm": 0.07518219947814941, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.8368, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0065506847364298765, | |
| "grad_norm": 0.06517896801233292, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.8774, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006835497116274654, | |
| "grad_norm": 0.0631726086139679, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.8199, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006835497116274654, | |
| "eval_loss": 0.8823266625404358, | |
| "eval_runtime": 301.5163, | |
| "eval_samples_per_second": 58.836, | |
| "eval_steps_per_second": 7.356, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007120309496119431, | |
| "grad_norm": 0.05915761739015579, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.7723, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.007405121875964209, | |
| "grad_norm": 0.0682271420955658, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.8155, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.007689934255808986, | |
| "grad_norm": 0.08482864499092102, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.8798, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007689934255808986, | |
| "eval_loss": 0.8814074397087097, | |
| "eval_runtime": 301.5821, | |
| "eval_samples_per_second": 58.823, | |
| "eval_steps_per_second": 7.355, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007974746635653764, | |
| "grad_norm": 0.06951498985290527, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.8692, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.00825955901549854, | |
| "grad_norm": 0.08030667901039124, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.9202, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.008544371395343317, | |
| "grad_norm": 0.08081574738025665, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.8877, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008544371395343317, | |
| "eval_loss": 0.8810605406761169, | |
| "eval_runtime": 301.1958, | |
| "eval_samples_per_second": 58.899, | |
| "eval_steps_per_second": 7.364, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008829183775188095, | |
| "grad_norm": 0.06468357145786285, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.8782, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.009113996155032873, | |
| "grad_norm": 0.07907775044441223, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.8721, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.009398808534877649, | |
| "grad_norm": 0.07169976830482483, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.8802, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.009398808534877649, | |
| "eval_loss": 0.8809642791748047, | |
| "eval_runtime": 301.1385, | |
| "eval_samples_per_second": 58.91, | |
| "eval_steps_per_second": 7.365, | |
| "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": 7898205577543680.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |