| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5563282336578581, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011126564673157162, | |
| "grad_norm": 0.45852020382881165, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7788, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011126564673157162, | |
| "eval_loss": 0.8189165592193604, | |
| "eval_runtime": 32.0686, | |
| "eval_samples_per_second": 4.74, | |
| "eval_steps_per_second": 2.37, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.022253129346314324, | |
| "grad_norm": 0.3102879822254181, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7235, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03337969401947149, | |
| "grad_norm": 0.4298129081726074, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6954, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04450625869262865, | |
| "grad_norm": 0.4370204508304596, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7928, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.055632823365785816, | |
| "grad_norm": 0.34161755442619324, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8569, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06675938803894298, | |
| "grad_norm": 0.2788660228252411, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7285, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07788595271210014, | |
| "grad_norm": 0.31495457887649536, | |
| "learning_rate": 7e-05, | |
| "loss": 0.7721, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0890125173852573, | |
| "grad_norm": 0.24107243120670319, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7058, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.10013908205841446, | |
| "grad_norm": 0.2623262107372284, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7499, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11126564673157163, | |
| "grad_norm": 0.30620405077934265, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6713, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12239221140472879, | |
| "grad_norm": 0.9073506593704224, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.748, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.13351877607788595, | |
| "grad_norm": 0.3162141740322113, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.8322, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.14464534075104313, | |
| "grad_norm": 0.2846090793609619, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.7256, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.14464534075104313, | |
| "eval_loss": 0.7517489194869995, | |
| "eval_runtime": 21.8923, | |
| "eval_samples_per_second": 6.943, | |
| "eval_steps_per_second": 3.472, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.15577190542420027, | |
| "grad_norm": 0.507627010345459, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.6689, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.16689847009735745, | |
| "grad_norm": 0.3133050203323364, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.6608, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1780250347705146, | |
| "grad_norm": 0.6398417949676514, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.7098, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18915159944367177, | |
| "grad_norm": 0.38771340250968933, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.6498, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.20027816411682892, | |
| "grad_norm": 0.3491506278514862, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7382, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2114047287899861, | |
| "grad_norm": 0.4874832034111023, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.6976, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.22253129346314326, | |
| "grad_norm": 0.4010584354400635, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7317, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2336578581363004, | |
| "grad_norm": 0.4318137466907501, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.6688, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.24478442280945759, | |
| "grad_norm": 0.26157474517822266, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.6209, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.25591098748261476, | |
| "grad_norm": 0.3200092613697052, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.7758, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2670375521557719, | |
| "grad_norm": 0.2472483515739441, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.7097, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.27816411682892905, | |
| "grad_norm": 0.44432780146598816, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.6646, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.28929068150208626, | |
| "grad_norm": 0.3959880769252777, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.7104, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.28929068150208626, | |
| "eval_loss": 0.7206902503967285, | |
| "eval_runtime": 21.7904, | |
| "eval_samples_per_second": 6.976, | |
| "eval_steps_per_second": 3.488, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3004172461752434, | |
| "grad_norm": 0.2234642058610916, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.6434, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.31154381084840055, | |
| "grad_norm": 0.28918102383613586, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.6914, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.3226703755215577, | |
| "grad_norm": 0.211372509598732, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.6567, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.3337969401947149, | |
| "grad_norm": 0.275553822517395, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5738, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.34492350486787204, | |
| "grad_norm": 0.34773099422454834, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.7286, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3560500695410292, | |
| "grad_norm": 0.22577843070030212, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.6837, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3671766342141864, | |
| "grad_norm": 0.29821667075157166, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.7444, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.37830319888734354, | |
| "grad_norm": 0.23734945058822632, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.627, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3894297635605007, | |
| "grad_norm": 0.22903797030448914, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.8233, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.40055632823365783, | |
| "grad_norm": 0.29000309109687805, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.5556, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.41168289290681503, | |
| "grad_norm": 0.2623641788959503, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.7741, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.4228094575799722, | |
| "grad_norm": 0.21798746287822723, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.6794, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.4339360222531293, | |
| "grad_norm": 0.28965574502944946, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.6445, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4339360222531293, | |
| "eval_loss": 0.7118078470230103, | |
| "eval_runtime": 21.7647, | |
| "eval_samples_per_second": 6.984, | |
| "eval_steps_per_second": 3.492, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.44506258692628653, | |
| "grad_norm": 0.26844772696495056, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.637, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4561891515994437, | |
| "grad_norm": 0.3411906361579895, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.6441, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.4673157162726008, | |
| "grad_norm": 0.3042059540748596, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.584, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.47844228094575797, | |
| "grad_norm": 0.24161170423030853, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.653, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.48956884561891517, | |
| "grad_norm": 0.23014530539512634, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.6039, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.5006954102920723, | |
| "grad_norm": 0.23123666644096375, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.6084, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5118219749652295, | |
| "grad_norm": 0.24543774127960205, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.67, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5229485396383866, | |
| "grad_norm": 0.2833655774593353, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.7551, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5340751043115438, | |
| "grad_norm": 0.3481193780899048, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.6205, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.545201668984701, | |
| "grad_norm": 0.3575291037559509, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.6147, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5563282336578581, | |
| "grad_norm": 0.3505605459213257, | |
| "learning_rate": 0.0, | |
| "loss": 0.6294, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.967549134241792e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |