| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0946408209806158, | |
| "eval_steps": 8, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012162675788673508, | |
| "eval_loss": 1.9050604104995728, | |
| "eval_runtime": 37.3442, | |
| "eval_samples_per_second": 7.417, | |
| "eval_steps_per_second": 3.722, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.036488027366020526, | |
| "grad_norm": 1.2019902467727661, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8705, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07297605473204105, | |
| "grad_norm": 1.1034138202667236, | |
| "learning_rate": 6e-05, | |
| "loss": 1.7926, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09730140630938806, | |
| "eval_loss": 1.5952084064483643, | |
| "eval_runtime": 36.9049, | |
| "eval_samples_per_second": 7.506, | |
| "eval_steps_per_second": 3.766, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.10946408209806158, | |
| "grad_norm": 1.1641662120819092, | |
| "learning_rate": 9e-05, | |
| "loss": 1.5738, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1459521094640821, | |
| "grad_norm": 1.0015640258789062, | |
| "learning_rate": 0.00012, | |
| "loss": 1.486, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18244013683010263, | |
| "grad_norm": 0.7985835075378418, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.4581, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.19460281261877613, | |
| "eval_loss": 1.405150294303894, | |
| "eval_runtime": 36.8992, | |
| "eval_samples_per_second": 7.507, | |
| "eval_steps_per_second": 3.767, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.21892816419612315, | |
| "grad_norm": 0.6988704204559326, | |
| "learning_rate": 0.00018, | |
| "loss": 1.4166, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2554161915621437, | |
| "grad_norm": 0.7365347743034363, | |
| "learning_rate": 0.00019989930665413147, | |
| "loss": 1.3802, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2919042189281642, | |
| "grad_norm": 0.767153799533844, | |
| "learning_rate": 0.00019839295885986296, | |
| "loss": 1.3818, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2919042189281642, | |
| "eval_loss": 1.3417332172393799, | |
| "eval_runtime": 36.8556, | |
| "eval_samples_per_second": 7.516, | |
| "eval_steps_per_second": 3.771, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.32839224629418473, | |
| "grad_norm": 0.8267173171043396, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 1.3572, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.36488027366020526, | |
| "grad_norm": 0.7445698380470276, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 1.2588, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.38920562523755226, | |
| "eval_loss": 1.305761694908142, | |
| "eval_runtime": 36.8956, | |
| "eval_samples_per_second": 7.508, | |
| "eval_steps_per_second": 3.767, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4013683010262258, | |
| "grad_norm": 0.6589832901954651, | |
| "learning_rate": 0.00018345732537213027, | |
| "loss": 1.3104, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4378563283922463, | |
| "grad_norm": 0.7035363912582397, | |
| "learning_rate": 0.00017530714660036112, | |
| "loss": 1.2935, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.47434435575826683, | |
| "grad_norm": 0.6706580519676208, | |
| "learning_rate": 0.00016579387259397127, | |
| "loss": 1.2888, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4865070315469403, | |
| "eval_loss": 1.286447286605835, | |
| "eval_runtime": 36.8997, | |
| "eval_samples_per_second": 7.507, | |
| "eval_steps_per_second": 3.767, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5108323831242874, | |
| "grad_norm": 0.6659970283508301, | |
| "learning_rate": 0.00015508969814521025, | |
| "loss": 1.2785, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5473204104903079, | |
| "grad_norm": 0.6708734035491943, | |
| "learning_rate": 0.00014338837391175582, | |
| "loss": 1.2349, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5838084378563284, | |
| "grad_norm": 0.6934670805931091, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 1.2415, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5838084378563284, | |
| "eval_loss": 1.271263599395752, | |
| "eval_runtime": 36.9007, | |
| "eval_samples_per_second": 7.507, | |
| "eval_steps_per_second": 3.767, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6202964652223489, | |
| "grad_norm": 0.6565465927124023, | |
| "learning_rate": 0.00011785568947986367, | |
| "loss": 1.2442, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6567844925883695, | |
| "grad_norm": 0.737869918346405, | |
| "learning_rate": 0.00010448648303505151, | |
| "loss": 1.2828, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6811098441657164, | |
| "eval_loss": 1.255753993988037, | |
| "eval_runtime": 36.8824, | |
| "eval_samples_per_second": 7.51, | |
| "eval_steps_per_second": 3.769, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.69327251995439, | |
| "grad_norm": 0.6686636209487915, | |
| "learning_rate": 9.103606910965666e-05, | |
| "loss": 1.1743, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.7297605473204105, | |
| "grad_norm": 0.720510721206665, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 1.2735, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.766248574686431, | |
| "grad_norm": 0.6112392544746399, | |
| "learning_rate": 6.486251759186572e-05, | |
| "loss": 1.2145, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7784112504751045, | |
| "eval_loss": 1.2483785152435303, | |
| "eval_runtime": 36.8553, | |
| "eval_samples_per_second": 7.516, | |
| "eval_steps_per_second": 3.772, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.8027366020524516, | |
| "grad_norm": 0.6711891293525696, | |
| "learning_rate": 5.261313375270014e-05, | |
| "loss": 1.2266, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.8392246294184721, | |
| "grad_norm": 0.6170798540115356, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 1.2298, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8757126567844926, | |
| "grad_norm": 0.6194135546684265, | |
| "learning_rate": 3.089373510131354e-05, | |
| "loss": 1.2306, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8757126567844926, | |
| "eval_loss": 1.239687204360962, | |
| "eval_runtime": 36.9026, | |
| "eval_samples_per_second": 7.506, | |
| "eval_steps_per_second": 3.767, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9122006841505131, | |
| "grad_norm": 0.6713999509811401, | |
| "learning_rate": 2.181685175319702e-05, | |
| "loss": 1.2185, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9486887115165337, | |
| "grad_norm": 0.6280702352523804, | |
| "learning_rate": 1.415512063981339e-05, | |
| "loss": 1.2253, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9730140630938806, | |
| "eval_loss": 1.2348722219467163, | |
| "eval_runtime": 36.9083, | |
| "eval_samples_per_second": 7.505, | |
| "eval_steps_per_second": 3.766, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9851767388825542, | |
| "grad_norm": 0.6385733485221863, | |
| "learning_rate": 8.047222744854943e-06, | |
| "loss": 1.2547, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0216647662485747, | |
| "grad_norm": 0.5922430157661438, | |
| "learning_rate": 3.6037139304146762e-06, | |
| "loss": 1.4369, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.0581527936145951, | |
| "grad_norm": 0.5931403040885925, | |
| "learning_rate": 9.0502382320653e-07, | |
| "loss": 1.0506, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.0703154694032686, | |
| "eval_loss": 1.2342017889022827, | |
| "eval_runtime": 36.8829, | |
| "eval_samples_per_second": 7.51, | |
| "eval_steps_per_second": 3.769, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.0946408209806158, | |
| "grad_norm": 0.5633111000061035, | |
| "learning_rate": 0.0, | |
| "loss": 1.0277, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.275812928834765e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |