| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.021276595744681, | |
| "eval_steps": 25, | |
| "global_step": 71, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0425531914893617, | |
| "grad_norm": 2.0995755195617676, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.1063, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0425531914893617, | |
| "eval_loss": 3.8875327110290527, | |
| "eval_runtime": 0.6786, | |
| "eval_samples_per_second": 29.471, | |
| "eval_steps_per_second": 14.735, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0851063829787234, | |
| "grad_norm": 3.263718605041504, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.4322, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1276595744680851, | |
| "grad_norm": 3.820951223373413, | |
| "learning_rate": 0.0001, | |
| "loss": 3.645, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1702127659574468, | |
| "grad_norm": 5.33889627456665, | |
| "learning_rate": 9.994664874011863e-05, | |
| "loss": 4.513, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 7.883951663970947, | |
| "learning_rate": 9.978670881475172e-05, | |
| "loss": 5.0264, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "grad_norm": 2.583144187927246, | |
| "learning_rate": 9.952052154376026e-05, | |
| "loss": 3.26, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2978723404255319, | |
| "grad_norm": 3.3901255130767822, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 3.2384, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 4.286656856536865, | |
| "learning_rate": 9.867190271803465e-05, | |
| "loss": 3.4462, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3829787234042553, | |
| "grad_norm": 4.722428321838379, | |
| "learning_rate": 9.809128215864097e-05, | |
| "loss": 3.7113, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 6.825965404510498, | |
| "learning_rate": 9.74080323795483e-05, | |
| "loss": 4.5429, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.46808510638297873, | |
| "grad_norm": 3.327223777770996, | |
| "learning_rate": 9.662361147021779e-05, | |
| "loss": 3.0301, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "grad_norm": 3.6547610759735107, | |
| "learning_rate": 9.573969342440106e-05, | |
| "loss": 2.9293, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5531914893617021, | |
| "grad_norm": 4.506838798522949, | |
| "learning_rate": 9.475816456775313e-05, | |
| "loss": 3.4406, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5957446808510638, | |
| "grad_norm": 5.095069408416748, | |
| "learning_rate": 9.368111953231848e-05, | |
| "loss": 3.5664, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 5.679275989532471, | |
| "learning_rate": 9.251085678648072e-05, | |
| "loss": 3.443, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 2.384618043899536, | |
| "learning_rate": 9.124987372991511e-05, | |
| "loss": 3.0016, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.723404255319149, | |
| "grad_norm": 3.2170469760894775, | |
| "learning_rate": 8.9900861364012e-05, | |
| "loss": 2.9207, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "grad_norm": 4.741713523864746, | |
| "learning_rate": 8.846669854914396e-05, | |
| "loss": 3.2638, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8085106382978723, | |
| "grad_norm": 4.597310543060303, | |
| "learning_rate": 8.695044586103296e-05, | |
| "loss": 3.119, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 6.106940746307373, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.196, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8936170212765957, | |
| "grad_norm": 3.007261037826538, | |
| "learning_rate": 8.368478218232787e-05, | |
| "loss": 2.7263, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.9361702127659575, | |
| "grad_norm": 3.8679449558258057, | |
| "learning_rate": 8.194234028259806e-05, | |
| "loss": 3.2128, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.9787234042553191, | |
| "grad_norm": 5.008350372314453, | |
| "learning_rate": 8.013173181896283e-05, | |
| "loss": 3.1513, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0212765957446808, | |
| "grad_norm": 8.066261291503906, | |
| "learning_rate": 7.82568207211296e-05, | |
| "loss": 5.1511, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "grad_norm": 2.237522840499878, | |
| "learning_rate": 7.63216081438678e-05, | |
| "loss": 2.0345, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_loss": 2.8090415000915527, | |
| "eval_runtime": 0.6742, | |
| "eval_samples_per_second": 29.666, | |
| "eval_steps_per_second": 14.833, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.1063829787234043, | |
| "grad_norm": 3.5807905197143555, | |
| "learning_rate": 7.433022392834282e-05, | |
| "loss": 2.552, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.148936170212766, | |
| "grad_norm": 3.804152250289917, | |
| "learning_rate": 7.228691778882693e-05, | |
| "loss": 2.7359, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.1914893617021276, | |
| "grad_norm": 3.969632625579834, | |
| "learning_rate": 7.019605024359474e-05, | |
| "loss": 2.6821, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.2340425531914894, | |
| "grad_norm": 3.872260808944702, | |
| "learning_rate": 6.806208330935766e-05, | |
| "loss": 2.7963, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "grad_norm": 2.6370015144348145, | |
| "learning_rate": 6.588957097909508e-05, | |
| "loss": 2.4097, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3191489361702127, | |
| "grad_norm": 3.32016658782959, | |
| "learning_rate": 6.368314950360415e-05, | |
| "loss": 2.4904, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.3617021276595744, | |
| "grad_norm": 4.197410583496094, | |
| "learning_rate": 6.14475274975067e-05, | |
| "loss": 2.5286, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.4042553191489362, | |
| "grad_norm": 5.017968654632568, | |
| "learning_rate": 5.918747589082853e-05, | |
| "loss": 2.9064, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.4468085106382977, | |
| "grad_norm": 3.416367292404175, | |
| "learning_rate": 5.6907817747594116e-05, | |
| "loss": 2.7989, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.4893617021276595, | |
| "grad_norm": 2.9843363761901855, | |
| "learning_rate": 5.4613417973165106e-05, | |
| "loss": 2.2761, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.5319148936170213, | |
| "grad_norm": 3.322655200958252, | |
| "learning_rate": 5.230917293228699e-05, | |
| "loss": 2.5173, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.574468085106383, | |
| "grad_norm": 4.156716823577881, | |
| "learning_rate": 5e-05, | |
| "loss": 2.6914, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.6170212765957448, | |
| "grad_norm": 5.216977596282959, | |
| "learning_rate": 4.7690827067713035e-05, | |
| "loss": 3.1488, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.6595744680851063, | |
| "grad_norm": 3.87528133392334, | |
| "learning_rate": 4.5386582026834906e-05, | |
| "loss": 2.8607, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 2.6391236782073975, | |
| "learning_rate": 4.30921822524059e-05, | |
| "loss": 2.315, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.7446808510638299, | |
| "grad_norm": 3.2835886478424072, | |
| "learning_rate": 4.0812524109171476e-05, | |
| "loss": 2.296, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.7872340425531914, | |
| "grad_norm": 4.525584697723389, | |
| "learning_rate": 3.855247250249331e-05, | |
| "loss": 2.9909, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.8297872340425532, | |
| "grad_norm": 5.08241605758667, | |
| "learning_rate": 3.631685049639586e-05, | |
| "loss": 2.5649, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.872340425531915, | |
| "grad_norm": 3.8431763648986816, | |
| "learning_rate": 3.411042902090492e-05, | |
| "loss": 2.9495, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.9148936170212765, | |
| "grad_norm": 2.8185856342315674, | |
| "learning_rate": 3.1937916690642356e-05, | |
| "loss": 2.326, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.9574468085106385, | |
| "grad_norm": 3.852102518081665, | |
| "learning_rate": 2.980394975640526e-05, | |
| "loss": 2.4781, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 9.235259056091309, | |
| "learning_rate": 2.771308221117309e-05, | |
| "loss": 4.1493, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.0425531914893615, | |
| "grad_norm": 2.182204484939575, | |
| "learning_rate": 2.5669776071657192e-05, | |
| "loss": 2.431, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.0851063829787235, | |
| "grad_norm": 2.5565974712371826, | |
| "learning_rate": 2.3678391856132204e-05, | |
| "loss": 2.2611, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "grad_norm": 3.7425453662872314, | |
| "learning_rate": 2.1743179278870407e-05, | |
| "loss": 2.5037, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 2.8088538646698, | |
| "eval_runtime": 0.6893, | |
| "eval_samples_per_second": 29.014, | |
| "eval_steps_per_second": 14.507, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.1702127659574466, | |
| "grad_norm": 4.252004623413086, | |
| "learning_rate": 1.9868268181037185e-05, | |
| "loss": 2.5753, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.2127659574468086, | |
| "grad_norm": 4.096494197845459, | |
| "learning_rate": 1.8057659717401947e-05, | |
| "loss": 2.25, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.25531914893617, | |
| "grad_norm": 2.2319579124450684, | |
| "learning_rate": 1.631521781767214e-05, | |
| "loss": 2.3128, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.297872340425532, | |
| "grad_norm": 2.539842367172241, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.2327, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.3404255319148937, | |
| "grad_norm": 3.147387981414795, | |
| "learning_rate": 1.3049554138967051e-05, | |
| "loss": 2.1849, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.382978723404255, | |
| "grad_norm": 3.685150146484375, | |
| "learning_rate": 1.1533301450856054e-05, | |
| "loss": 2.4277, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.425531914893617, | |
| "grad_norm": 4.715639591217041, | |
| "learning_rate": 1.0099138635988026e-05, | |
| "loss": 2.5731, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.4680851063829787, | |
| "grad_norm": 2.3089892864227295, | |
| "learning_rate": 8.75012627008489e-06, | |
| "loss": 2.4525, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.5106382978723403, | |
| "grad_norm": 2.630906105041504, | |
| "learning_rate": 7.489143213519301e-06, | |
| "loss": 2.1467, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 3.5077598094940186, | |
| "learning_rate": 6.318880467681526e-06, | |
| "loss": 2.4428, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.595744680851064, | |
| "grad_norm": 3.928353786468506, | |
| "learning_rate": 5.241835432246889e-06, | |
| "loss": 2.2675, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.6382978723404253, | |
| "grad_norm": 4.511261940002441, | |
| "learning_rate": 4.260306575598949e-06, | |
| "loss": 2.2792, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.6808510638297873, | |
| "grad_norm": 2.4075710773468018, | |
| "learning_rate": 3.376388529782215e-06, | |
| "loss": 2.3584, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.723404255319149, | |
| "grad_norm": 2.83492112159729, | |
| "learning_rate": 2.591967620451707e-06, | |
| "loss": 1.9989, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.7659574468085104, | |
| "grad_norm": 3.5723989009857178, | |
| "learning_rate": 1.908717841359048e-06, | |
| "loss": 2.2559, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.8085106382978724, | |
| "grad_norm": 3.7944796085357666, | |
| "learning_rate": 1.328097281965357e-06, | |
| "loss": 2.2629, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.851063829787234, | |
| "grad_norm": 4.501901626586914, | |
| "learning_rate": 8.513450158049108e-07, | |
| "loss": 2.3413, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.8936170212765955, | |
| "grad_norm": 2.496269464492798, | |
| "learning_rate": 4.794784562397458e-07, | |
| "loss": 2.2531, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.9361702127659575, | |
| "grad_norm": 2.9496850967407227, | |
| "learning_rate": 2.1329118524827662e-07, | |
| "loss": 2.1338, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.978723404255319, | |
| "grad_norm": 4.252143383026123, | |
| "learning_rate": 5.3351259881379014e-08, | |
| "loss": 2.4166, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.021276595744681, | |
| "grad_norm": 7.078902721405029, | |
| "learning_rate": 0.0, | |
| "loss": 4.2177, | |
| "step": 71 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 71, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3899010909667328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |