| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0033823921122615942, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.764784224523189e-05, | |
| "grad_norm": 5.634762287139893, | |
| "learning_rate": 1e-05, | |
| "loss": 3.0597, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.764784224523189e-05, | |
| "eval_loss": 3.2875728607177734, | |
| "eval_runtime": 948.466, | |
| "eval_samples_per_second": 26.25, | |
| "eval_steps_per_second": 13.125, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00013529568449046378, | |
| "grad_norm": 5.37744140625, | |
| "learning_rate": 2e-05, | |
| "loss": 3.3157, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00020294352673569564, | |
| "grad_norm": 2.809889078140259, | |
| "learning_rate": 3e-05, | |
| "loss": 3.2284, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00027059136898092756, | |
| "grad_norm": 1.9245902299880981, | |
| "learning_rate": 4e-05, | |
| "loss": 2.5244, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00033823921122615945, | |
| "grad_norm": 1.7348881959915161, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2311, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004058870534713913, | |
| "grad_norm": 1.531990885734558, | |
| "learning_rate": 6e-05, | |
| "loss": 3.0822, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00047353489571662317, | |
| "grad_norm": 1.5458530187606812, | |
| "learning_rate": 7e-05, | |
| "loss": 3.1807, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0005411827379618551, | |
| "grad_norm": 1.52450692653656, | |
| "learning_rate": 8e-05, | |
| "loss": 2.732, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.000608830580207087, | |
| "grad_norm": 2.0526621341705322, | |
| "learning_rate": 9e-05, | |
| "loss": 2.5615, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006764784224523189, | |
| "grad_norm": 1.7713125944137573, | |
| "learning_rate": 0.0001, | |
| "loss": 2.4923, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007441262646975507, | |
| "grad_norm": 1.1165575981140137, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 2.8424, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0008117741069427826, | |
| "grad_norm": 2.138780355453491, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 2.3238, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008794219491880145, | |
| "grad_norm": 1.88950777053833, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 2.592, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0008794219491880145, | |
| "eval_loss": 1.9662768840789795, | |
| "eval_runtime": 872.1234, | |
| "eval_samples_per_second": 28.548, | |
| "eval_steps_per_second": 14.274, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0009470697914332463, | |
| "grad_norm": 2.2020058631896973, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.5448, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0010147176336784782, | |
| "grad_norm": 1.4748566150665283, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.622, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0010823654759237102, | |
| "grad_norm": 1.5124787092208862, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 2.6661, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.001150013318168942, | |
| "grad_norm": 2.291480541229248, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.9446, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.001217661160414174, | |
| "grad_norm": 1.5338326692581177, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.5786, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012853090026594058, | |
| "grad_norm": 1.017585039138794, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.669, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0013529568449046378, | |
| "grad_norm": 1.0846589803695679, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.8216, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014206046871498696, | |
| "grad_norm": 1.0249284505844116, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.1166, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0014882525293951013, | |
| "grad_norm": 1.2318507432937622, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.7656, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0015559003716403333, | |
| "grad_norm": 1.3429690599441528, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.8599, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0016235482138855651, | |
| "grad_norm": 0.7906981110572815, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.1856, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0016911960561307971, | |
| "grad_norm": 0.8192072510719299, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.9776, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001758843898376029, | |
| "grad_norm": 0.7536182403564453, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.8015, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.001758843898376029, | |
| "eval_loss": 1.444047451019287, | |
| "eval_runtime": 871.7811, | |
| "eval_samples_per_second": 28.559, | |
| "eval_steps_per_second": 14.28, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.001826491740621261, | |
| "grad_norm": 0.7033780813217163, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.8607, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0018941395828664927, | |
| "grad_norm": 0.8126703500747681, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.2693, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0019617874251117245, | |
| "grad_norm": 0.6983233690261841, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 2.1051, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0020294352673569565, | |
| "grad_norm": 0.6291278600692749, | |
| "learning_rate": 5e-05, | |
| "loss": 2.5043, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0020970831096021885, | |
| "grad_norm": 0.7705021500587463, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 2.0858, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0021647309518474205, | |
| "grad_norm": 0.7353219985961914, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 2.121, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.002232378794092652, | |
| "grad_norm": 0.7408473491668701, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.4857, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.002300026636337884, | |
| "grad_norm": 0.5902467966079712, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.047, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.002367674478583116, | |
| "grad_norm": 0.7522490620613098, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.6697, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.002435322320828348, | |
| "grad_norm": 0.7086387872695923, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.5194, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0025029701630735796, | |
| "grad_norm": 0.7631036043167114, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 2.1248, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0025706180053188116, | |
| "grad_norm": 0.5777090191841125, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.13, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0026382658475640436, | |
| "grad_norm": 0.7917817234992981, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.5179, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0026382658475640436, | |
| "eval_loss": 1.3922735452651978, | |
| "eval_runtime": 871.5053, | |
| "eval_samples_per_second": 28.568, | |
| "eval_steps_per_second": 14.284, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0027059136898092756, | |
| "grad_norm": 0.787195086479187, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.4687, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.002773561532054507, | |
| "grad_norm": 0.7193476557731628, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.8403, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.002841209374299739, | |
| "grad_norm": 0.7137141227722168, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.396, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.002908857216544971, | |
| "grad_norm": 0.571948230266571, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.6501, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0029765050587902027, | |
| "grad_norm": 0.5318587422370911, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.1225, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0030441529010354347, | |
| "grad_norm": 0.7348474860191345, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.4402, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0031118007432806667, | |
| "grad_norm": 0.7427931427955627, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.9161, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0031794485855258987, | |
| "grad_norm": 0.917795717716217, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 2.0163, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0032470964277711302, | |
| "grad_norm": 0.6537682414054871, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.3718, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0033147442700163622, | |
| "grad_norm": 0.796440839767456, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.6624, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0033823921122615942, | |
| "grad_norm": 0.7888891696929932, | |
| "learning_rate": 0.0, | |
| "loss": 1.7127, | |
| "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": 1.4419086016512e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |