| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12277470841006753, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0012277470841006752, | |
| "eval_loss": 0.8551988005638123, | |
| "eval_runtime": 202.8446, | |
| "eval_samples_per_second": 6.764, | |
| "eval_steps_per_second": 0.848, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003683241252302026, | |
| "grad_norm": 5.180573463439941, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.3333, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007366482504604052, | |
| "grad_norm": 3.8857433795928955, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4064, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011049723756906077, | |
| "grad_norm": 2.247364044189453, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.0399, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011049723756906077, | |
| "eval_loss": 0.7726849913597107, | |
| "eval_runtime": 205.2481, | |
| "eval_samples_per_second": 6.685, | |
| "eval_steps_per_second": 0.838, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014732965009208104, | |
| "grad_norm": 2.557511568069458, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.3922, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01841620626151013, | |
| "grad_norm": 2.2613649368286133, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.6714, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022099447513812154, | |
| "grad_norm": 2.2691195011138916, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.8087, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.022099447513812154, | |
| "eval_loss": 0.7028216123580933, | |
| "eval_runtime": 205.2466, | |
| "eval_samples_per_second": 6.685, | |
| "eval_steps_per_second": 0.838, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02578268876611418, | |
| "grad_norm": 2.2336041927337646, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 2.8286, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.029465930018416207, | |
| "grad_norm": 2.087846040725708, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.7683, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03314917127071823, | |
| "grad_norm": 2.1775248050689697, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.7827, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03314917127071823, | |
| "eval_loss": 0.6572425961494446, | |
| "eval_runtime": 205.2001, | |
| "eval_samples_per_second": 6.686, | |
| "eval_steps_per_second": 0.838, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03683241252302026, | |
| "grad_norm": 1.9543025493621826, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.5388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.040515653775322284, | |
| "grad_norm": 2.1249802112579346, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.6073, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04419889502762431, | |
| "grad_norm": 2.4751439094543457, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.5199, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04419889502762431, | |
| "eval_loss": 0.6265907287597656, | |
| "eval_runtime": 205.1795, | |
| "eval_samples_per_second": 6.687, | |
| "eval_steps_per_second": 0.838, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04788213627992634, | |
| "grad_norm": 2.063572645187378, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.4584, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05156537753222836, | |
| "grad_norm": 2.1027767658233643, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.2569, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.055248618784530384, | |
| "grad_norm": 2.1537835597991943, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.4282, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.055248618784530384, | |
| "eval_loss": 0.6082205772399902, | |
| "eval_runtime": 205.0067, | |
| "eval_samples_per_second": 6.692, | |
| "eval_steps_per_second": 0.839, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.058931860036832415, | |
| "grad_norm": 1.9434477090835571, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.4655, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06261510128913444, | |
| "grad_norm": 2.0563454627990723, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.3624, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.06629834254143646, | |
| "grad_norm": 2.1307311058044434, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.4587, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06629834254143646, | |
| "eval_loss": 0.5953105092048645, | |
| "eval_runtime": 205.3342, | |
| "eval_samples_per_second": 6.682, | |
| "eval_steps_per_second": 0.838, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06998158379373849, | |
| "grad_norm": 2.182778835296631, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.6524, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.07366482504604052, | |
| "grad_norm": 2.055069923400879, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.0877, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07734806629834254, | |
| "grad_norm": 2.2690212726593018, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.5206, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.07734806629834254, | |
| "eval_loss": 0.586376965045929, | |
| "eval_runtime": 205.3086, | |
| "eval_samples_per_second": 6.683, | |
| "eval_steps_per_second": 0.838, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08103130755064457, | |
| "grad_norm": 2.091456174850464, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.3057, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0847145488029466, | |
| "grad_norm": 2.572800397872925, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.5019, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.08839779005524862, | |
| "grad_norm": 2.008159875869751, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.3352, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08839779005524862, | |
| "eval_loss": 0.5802063345909119, | |
| "eval_runtime": 205.308, | |
| "eval_samples_per_second": 6.683, | |
| "eval_steps_per_second": 0.838, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09208103130755065, | |
| "grad_norm": 2.07601261138916, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.3431, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.09576427255985268, | |
| "grad_norm": 1.9664666652679443, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.3686, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.09944751381215469, | |
| "grad_norm": 2.1633036136627197, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.4072, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.09944751381215469, | |
| "eval_loss": 0.5766007900238037, | |
| "eval_runtime": 205.4043, | |
| "eval_samples_per_second": 6.68, | |
| "eval_steps_per_second": 0.837, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.10313075506445672, | |
| "grad_norm": 2.327897071838379, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.3151, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.10681399631675875, | |
| "grad_norm": 2.321183919906616, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.4353, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.11049723756906077, | |
| "grad_norm": 2.2295124530792236, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.3415, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11049723756906077, | |
| "eval_loss": 0.5750809907913208, | |
| "eval_runtime": 205.4722, | |
| "eval_samples_per_second": 6.677, | |
| "eval_steps_per_second": 0.837, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1141804788213628, | |
| "grad_norm": 2.291185140609741, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.1224, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.11786372007366483, | |
| "grad_norm": 2.3587148189544678, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.3583, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.12154696132596685, | |
| "grad_norm": 2.1020097732543945, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.1685, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.12154696132596685, | |
| "eval_loss": 0.574731171131134, | |
| "eval_runtime": 205.3596, | |
| "eval_samples_per_second": 6.681, | |
| "eval_steps_per_second": 0.838, | |
| "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": 2.812522826563584e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |