| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.019984733883838736, | |
| "eval_steps": 9, | |
| "global_step": 72, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002775657483866491, | |
| "eval_loss": 0.5042526125907898, | |
| "eval_runtime": 655.4432, | |
| "eval_samples_per_second": 9.258, | |
| "eval_steps_per_second": 1.158, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008326972451599472, | |
| "grad_norm": 0.27662014961242676, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.526, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0016653944903198944, | |
| "grad_norm": 0.27963921427726746, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4897, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002498091735479842, | |
| "grad_norm": 0.29799002408981323, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5052, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002498091735479842, | |
| "eval_loss": 0.4782797694206238, | |
| "eval_runtime": 660.6166, | |
| "eval_samples_per_second": 9.185, | |
| "eval_steps_per_second": 1.149, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003330788980639789, | |
| "grad_norm": 0.23466083407402039, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.4373, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004163486225799737, | |
| "grad_norm": 0.20258335769176483, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.4476, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004996183470959684, | |
| "grad_norm": 0.22826525568962097, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.4442, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004996183470959684, | |
| "eval_loss": 0.42058902978897095, | |
| "eval_runtime": 660.3378, | |
| "eval_samples_per_second": 9.189, | |
| "eval_steps_per_second": 1.149, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0058288807161196304, | |
| "grad_norm": 0.20002856850624084, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.4395, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006661577961279578, | |
| "grad_norm": 0.21906918287277222, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.423, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007494275206439525, | |
| "grad_norm": 0.1747865527868271, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.4055, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007494275206439525, | |
| "eval_loss": 0.39028802514076233, | |
| "eval_runtime": 660.3218, | |
| "eval_samples_per_second": 9.189, | |
| "eval_steps_per_second": 1.149, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008326972451599473, | |
| "grad_norm": 0.15783219039440155, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.3895, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00915966969675942, | |
| "grad_norm": 0.17900213599205017, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.4093, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.009992366941919368, | |
| "grad_norm": 0.14911402761936188, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.3543, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009992366941919368, | |
| "eval_loss": 0.3740505874156952, | |
| "eval_runtime": 660.2612, | |
| "eval_samples_per_second": 9.19, | |
| "eval_steps_per_second": 1.15, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010825064187079314, | |
| "grad_norm": 0.19286595284938812, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.378, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.011657761432239261, | |
| "grad_norm": 0.20762427151203156, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.4039, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.012490458677399209, | |
| "grad_norm": 0.1565249115228653, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.3701, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.012490458677399209, | |
| "eval_loss": 0.36458876729011536, | |
| "eval_runtime": 659.9797, | |
| "eval_samples_per_second": 9.194, | |
| "eval_steps_per_second": 1.15, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013323155922559156, | |
| "grad_norm": 0.1587727814912796, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.363, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.014155853167719104, | |
| "grad_norm": 0.19063134491443634, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.3633, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01498855041287905, | |
| "grad_norm": 0.16761146485805511, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.3519, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01498855041287905, | |
| "eval_loss": 0.3595520257949829, | |
| "eval_runtime": 660.1365, | |
| "eval_samples_per_second": 9.192, | |
| "eval_steps_per_second": 1.15, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.015821247658039, | |
| "grad_norm": 0.17183250188827515, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.3577, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.016653944903198947, | |
| "grad_norm": 0.18601562082767487, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.3407, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01748664214835889, | |
| "grad_norm": 0.17450034618377686, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.3614, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01748664214835889, | |
| "eval_loss": 0.35599005222320557, | |
| "eval_runtime": 659.9247, | |
| "eval_samples_per_second": 9.195, | |
| "eval_steps_per_second": 1.15, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01831933939351884, | |
| "grad_norm": 0.1783250868320465, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.357, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.019152036638678788, | |
| "grad_norm": 0.1924220472574234, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.364, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.019984733883838736, | |
| "grad_norm": 0.1912061870098114, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.3565, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.019984733883838736, | |
| "eval_loss": 0.35400116443634033, | |
| "eval_runtime": 660.0962, | |
| "eval_samples_per_second": 9.193, | |
| "eval_steps_per_second": 1.15, | |
| "step": 72 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5693418665187738e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |