| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04877453969028167, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00048774539690281675, | |
| "eval_loss": 1.048367977142334, | |
| "eval_runtime": 282.0297, | |
| "eval_samples_per_second": 12.243, | |
| "eval_steps_per_second": 1.532, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014632361907084503, | |
| "grad_norm": 0.37770092487335205, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.0341, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0029264723814169005, | |
| "grad_norm": 0.4068027436733246, | |
| "learning_rate": 3e-05, | |
| "loss": 1.056, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0043897085721253505, | |
| "grad_norm": 0.5370364189147949, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.0776, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0043897085721253505, | |
| "eval_loss": 0.9989280104637146, | |
| "eval_runtime": 284.091, | |
| "eval_samples_per_second": 12.155, | |
| "eval_steps_per_second": 1.521, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005852944762833801, | |
| "grad_norm": 0.39467090368270874, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.9114, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007316180953542251, | |
| "grad_norm": 0.472395658493042, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.8932, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008779417144250701, | |
| "grad_norm": 0.4325377941131592, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.796, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008779417144250701, | |
| "eval_loss": 0.8529807329177856, | |
| "eval_runtime": 284.0503, | |
| "eval_samples_per_second": 12.156, | |
| "eval_steps_per_second": 1.521, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.010242653334959152, | |
| "grad_norm": 0.4993043541908264, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.8698, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.011705889525667602, | |
| "grad_norm": 0.4442359507083893, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.6909, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.013169125716376053, | |
| "grad_norm": 0.4680705666542053, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.8662, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.013169125716376053, | |
| "eval_loss": 0.8155685067176819, | |
| "eval_runtime": 284.0968, | |
| "eval_samples_per_second": 12.154, | |
| "eval_steps_per_second": 1.521, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.014632361907084501, | |
| "grad_norm": 0.6314910054206848, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.7935, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016095598097792953, | |
| "grad_norm": 0.3279048800468445, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.823, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.017558834288501402, | |
| "grad_norm": 0.4065856635570526, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.8136, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.017558834288501402, | |
| "eval_loss": 0.7961127161979675, | |
| "eval_runtime": 284.0262, | |
| "eval_samples_per_second": 12.157, | |
| "eval_steps_per_second": 1.521, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01902207047920985, | |
| "grad_norm": 0.3428895175457001, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.7474, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.020485306669918303, | |
| "grad_norm": 0.5276553630828857, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.8486, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.021948542860626752, | |
| "grad_norm": 0.45164918899536133, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.7817, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.021948542860626752, | |
| "eval_loss": 0.7856877446174622, | |
| "eval_runtime": 284.5888, | |
| "eval_samples_per_second": 12.133, | |
| "eval_steps_per_second": 1.518, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.023411779051335204, | |
| "grad_norm": 0.530996561050415, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.8087, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.024875015242043653, | |
| "grad_norm": 0.4066341519355774, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.757, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.026338251432752105, | |
| "grad_norm": 0.49550753831863403, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.8163, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.026338251432752105, | |
| "eval_loss": 0.7791209816932678, | |
| "eval_runtime": 284.3762, | |
| "eval_samples_per_second": 12.142, | |
| "eval_steps_per_second": 1.519, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.027801487623460554, | |
| "grad_norm": 0.37957167625427246, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.7212, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.029264723814169002, | |
| "grad_norm": 0.40554001927375793, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.7365, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.030727960004877455, | |
| "grad_norm": 0.5294321179389954, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.7684, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.030727960004877455, | |
| "eval_loss": 0.7754349112510681, | |
| "eval_runtime": 284.4634, | |
| "eval_samples_per_second": 12.139, | |
| "eval_steps_per_second": 1.519, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03219119619558591, | |
| "grad_norm": 0.6118860840797424, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.7783, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.033654432386294356, | |
| "grad_norm": 0.4474002718925476, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.7547, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.035117668577002804, | |
| "grad_norm": 0.5504525303840637, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.8348, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.035117668577002804, | |
| "eval_loss": 0.7724781632423401, | |
| "eval_runtime": 284.4582, | |
| "eval_samples_per_second": 12.139, | |
| "eval_steps_per_second": 1.519, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03658090476771125, | |
| "grad_norm": 0.4114806056022644, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.7906, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0380441409584197, | |
| "grad_norm": 0.49407699704170227, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.7882, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.03950737714912816, | |
| "grad_norm": 0.687836229801178, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.7283, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03950737714912816, | |
| "eval_loss": 0.7711643576622009, | |
| "eval_runtime": 284.6778, | |
| "eval_samples_per_second": 12.13, | |
| "eval_steps_per_second": 1.518, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.040970613339836606, | |
| "grad_norm": 0.38511985540390015, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.8136, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.042433849530545055, | |
| "grad_norm": 0.45352721214294434, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.8204, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.043897085721253504, | |
| "grad_norm": 0.32951414585113525, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.7356, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.043897085721253504, | |
| "eval_loss": 0.770723819732666, | |
| "eval_runtime": 284.6788, | |
| "eval_samples_per_second": 12.129, | |
| "eval_steps_per_second": 1.517, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04536032191196195, | |
| "grad_norm": 0.42109906673431396, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.7546, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.04682355810267041, | |
| "grad_norm": 0.47802457213401794, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.6932, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04828679429337886, | |
| "grad_norm": 0.36127352714538574, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.8052, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04828679429337886, | |
| "eval_loss": 0.7704710364341736, | |
| "eval_runtime": 284.3211, | |
| "eval_samples_per_second": 12.145, | |
| "eval_steps_per_second": 1.519, | |
| "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": 1.6637120901178982e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |