| { | |
| "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": 10.594181060791016, | |
| "eval_runtime": 113.8111, | |
| "eval_samples_per_second": 155.872, | |
| "eval_steps_per_second": 19.488, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002848123798447773, | |
| "grad_norm": 0.8785746097564697, | |
| "learning_rate": 1.5e-05, | |
| "loss": 10.5968, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005696247596895546, | |
| "grad_norm": 0.7744207382202148, | |
| "learning_rate": 3e-05, | |
| "loss": 10.5731, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008544371395343318, | |
| "grad_norm": 0.8253825306892395, | |
| "learning_rate": 4.5e-05, | |
| "loss": 10.5738, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008544371395343318, | |
| "eval_loss": 10.563873291015625, | |
| "eval_runtime": 113.9599, | |
| "eval_samples_per_second": 155.669, | |
| "eval_steps_per_second": 19.463, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011392495193791091, | |
| "grad_norm": 0.9217737913131714, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 10.5507, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0014240618992238862, | |
| "grad_norm": 1.0646398067474365, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 10.503, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0017088742790686636, | |
| "grad_norm": 1.061179757118225, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 10.4435, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0017088742790686636, | |
| "eval_loss": 10.36979866027832, | |
| "eval_runtime": 113.6826, | |
| "eval_samples_per_second": 156.049, | |
| "eval_steps_per_second": 19.51, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001993686658913441, | |
| "grad_norm": 1.328324556350708, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 10.3437, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0022784990387582182, | |
| "grad_norm": 1.2739237546920776, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 10.264, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002563311418602995, | |
| "grad_norm": 1.499740719795227, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 10.1468, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002563311418602995, | |
| "eval_loss": 10.06716537475586, | |
| "eval_runtime": 113.41, | |
| "eval_samples_per_second": 156.424, | |
| "eval_steps_per_second": 19.557, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0028481237984477725, | |
| "grad_norm": 1.6156394481658936, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 10.0428, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00313293617829255, | |
| "grad_norm": 1.7073334455490112, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 9.9061, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003417748558137327, | |
| "grad_norm": 1.6044425964355469, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 9.8311, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003417748558137327, | |
| "eval_loss": 9.738450050354004, | |
| "eval_runtime": 114.0416, | |
| "eval_samples_per_second": 155.557, | |
| "eval_steps_per_second": 19.449, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0037025609379821045, | |
| "grad_norm": 1.611886739730835, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 9.7576, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003987373317826882, | |
| "grad_norm": 1.7682589292526245, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 9.5977, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004272185697671659, | |
| "grad_norm": 1.7930411100387573, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 9.4899, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004272185697671659, | |
| "eval_loss": 9.44653034210205, | |
| "eval_runtime": 113.5607, | |
| "eval_samples_per_second": 156.216, | |
| "eval_steps_per_second": 19.531, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0045569980775164365, | |
| "grad_norm": 1.6953896284103394, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 9.4481, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004841810457361213, | |
| "grad_norm": 1.7754610776901245, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 9.363, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.00512662283720599, | |
| "grad_norm": 1.6957319974899292, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 9.2884, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.00512662283720599, | |
| "eval_loss": 9.21696949005127, | |
| "eval_runtime": 113.9372, | |
| "eval_samples_per_second": 155.7, | |
| "eval_steps_per_second": 19.467, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005411435217050768, | |
| "grad_norm": 1.9713038206100464, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 9.1375, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005696247596895545, | |
| "grad_norm": 1.8748242855072021, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 9.1021, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005981059976740323, | |
| "grad_norm": 1.7590270042419434, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 9.1081, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005981059976740323, | |
| "eval_loss": 9.055033683776855, | |
| "eval_runtime": 113.6017, | |
| "eval_samples_per_second": 156.16, | |
| "eval_steps_per_second": 19.524, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0062658723565851, | |
| "grad_norm": 1.6782749891281128, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 8.9914, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0065506847364298765, | |
| "grad_norm": 1.5772511959075928, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 9.0474, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006835497116274654, | |
| "grad_norm": 1.980194091796875, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 8.957, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006835497116274654, | |
| "eval_loss": 8.955224990844727, | |
| "eval_runtime": 114.5372, | |
| "eval_samples_per_second": 154.884, | |
| "eval_steps_per_second": 19.365, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007120309496119431, | |
| "grad_norm": 1.7712100744247437, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 8.9299, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.007405121875964209, | |
| "grad_norm": 1.6637734174728394, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 8.8943, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.007689934255808986, | |
| "grad_norm": 1.4766582250595093, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 8.9262, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007689934255808986, | |
| "eval_loss": 8.90304183959961, | |
| "eval_runtime": 113.7323, | |
| "eval_samples_per_second": 155.98, | |
| "eval_steps_per_second": 19.502, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007974746635653764, | |
| "grad_norm": 1.5637836456298828, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 8.8611, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.00825955901549854, | |
| "grad_norm": 1.6171281337738037, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 8.9338, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.008544371395343317, | |
| "grad_norm": 1.5863943099975586, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 8.9591, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008544371395343317, | |
| "eval_loss": 8.882512092590332, | |
| "eval_runtime": 113.4322, | |
| "eval_samples_per_second": 156.393, | |
| "eval_steps_per_second": 19.554, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008829183775188095, | |
| "grad_norm": 1.6321587562561035, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 8.9392, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.009113996155032873, | |
| "grad_norm": 1.5234466791152954, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 8.8882, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.009398808534877649, | |
| "grad_norm": 1.6211401224136353, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 8.8878, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.009398808534877649, | |
| "eval_loss": 8.878786087036133, | |
| "eval_runtime": 114.8285, | |
| "eval_samples_per_second": 154.491, | |
| "eval_steps_per_second": 19.316, | |
| "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": 8571248128819200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |