| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.4444444444444446, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.024691358024691357, | |
| "eval_loss": 2.1480655670166016, | |
| "eval_runtime": 1.0117, | |
| "eval_samples_per_second": 68.202, | |
| "eval_steps_per_second": 8.896, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 0.17830029129981995, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.2572, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 0.21468417346477509, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2267, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.22635942697525024, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.271, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "eval_loss": 2.143685817718506, | |
| "eval_runtime": 1.0061, | |
| "eval_samples_per_second": 68.58, | |
| "eval_steps_per_second": 8.945, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 0.1852104514837265, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.2203, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 0.24263235926628113, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.165, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.24341049790382385, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.2557, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "eval_loss": 2.1180965900421143, | |
| "eval_runtime": 1.0063, | |
| "eval_samples_per_second": 68.568, | |
| "eval_steps_per_second": 8.944, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5185185185185185, | |
| "grad_norm": 0.2104165405035019, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 2.2173, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.1979183554649353, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.1982, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.23016861081123352, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.2124, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "eval_loss": 2.072659730911255, | |
| "eval_runtime": 1.0059, | |
| "eval_samples_per_second": 68.595, | |
| "eval_steps_per_second": 8.947, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 0.14867456257343292, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.1073, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8148148148148148, | |
| "grad_norm": 0.20560601353645325, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.1231, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.20878741145133972, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.1412, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_loss": 2.027235746383667, | |
| "eval_runtime": 1.0148, | |
| "eval_samples_per_second": 67.995, | |
| "eval_steps_per_second": 8.869, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9629629629629629, | |
| "grad_norm": 0.24395641684532166, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.1021, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.037037037037037, | |
| "grad_norm": 0.20625323057174683, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.5074, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.23216880857944489, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.0798, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "eval_loss": 1.9839798212051392, | |
| "eval_runtime": 1.0128, | |
| "eval_samples_per_second": 68.128, | |
| "eval_steps_per_second": 8.886, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.20464646816253662, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.0576, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.2592592592592593, | |
| "grad_norm": 0.20127850770950317, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.9593, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.2141118198633194, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.1897, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "eval_loss": 1.9443175792694092, | |
| "eval_runtime": 1.0093, | |
| "eval_samples_per_second": 68.361, | |
| "eval_steps_per_second": 8.917, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.4074074074074074, | |
| "grad_norm": 0.31057700514793396, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.9642, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.4814814814814814, | |
| "grad_norm": 0.30521175265312195, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.0191, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 0.1954805850982666, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.1129, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "eval_loss": 1.9135476350784302, | |
| "eval_runtime": 1.0142, | |
| "eval_samples_per_second": 68.033, | |
| "eval_steps_per_second": 8.874, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.6296296296296298, | |
| "grad_norm": 0.18209777772426605, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.8804, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.7037037037037037, | |
| "grad_norm": 0.19758297502994537, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.8818, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.19676625728607178, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.0798, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "eval_loss": 1.9025213718414307, | |
| "eval_runtime": 1.0144, | |
| "eval_samples_per_second": 68.024, | |
| "eval_steps_per_second": 8.873, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.8518518518518519, | |
| "grad_norm": 0.24667955935001373, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.8911, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.925925925925926, | |
| "grad_norm": 0.20230413973331451, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.0511, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2845350503921509, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.298, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8977771997451782, | |
| "eval_runtime": 1.0089, | |
| "eval_samples_per_second": 68.393, | |
| "eval_steps_per_second": 8.921, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.074074074074074, | |
| "grad_norm": 0.20115576684474945, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.0542, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.148148148148148, | |
| "grad_norm": 0.1773402839899063, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.0123, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.20624512434005737, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.9569, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "eval_loss": 1.8960562944412231, | |
| "eval_runtime": 1.0098, | |
| "eval_samples_per_second": 68.33, | |
| "eval_steps_per_second": 8.913, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.2962962962962963, | |
| "grad_norm": 0.20254096388816833, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.8373, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.20453697443008423, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.9392, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "grad_norm": 0.1803455650806427, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.0326, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.4444444444444446, | |
| "eval_loss": 1.8957442045211792, | |
| "eval_runtime": 1.0119, | |
| "eval_samples_per_second": 68.192, | |
| "eval_steps_per_second": 8.895, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6289164702842880.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |