| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1208, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.026490066225165563, | |
| "grad_norm": 0.13324208557605743, | |
| "learning_rate": 0.0002191780821917808, | |
| "loss": 4.2964, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.052980132450331126, | |
| "grad_norm": 0.1445072889328003, | |
| "learning_rate": 0.0004383561643835616, | |
| "loss": 4.2649, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.07947019867549669, | |
| "grad_norm": 0.14529070258140564, | |
| "learning_rate": 0.0006575342465753425, | |
| "loss": 4.3014, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.10596026490066225, | |
| "grad_norm": 0.2007322758436203, | |
| "learning_rate": 0.0008767123287671232, | |
| "loss": 4.3097, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.13245033112582782, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0009999959548269918, | |
| "loss": 4.3854, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15894039735099338, | |
| "grad_norm": 0.2345261424779892, | |
| "learning_rate": 0.0009998377521727646, | |
| "loss": 4.4083, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.18543046357615894, | |
| "grad_norm": 0.14606653153896332, | |
| "learning_rate": 0.0009994495073085382, | |
| "loss": 4.3126, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2119205298013245, | |
| "grad_norm": 0.12169423699378967, | |
| "learning_rate": 0.0009988313989177076, | |
| "loss": 4.2892, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.23841059602649006, | |
| "grad_norm": 0.21712887287139893, | |
| "learning_rate": 0.0009979837114746243, | |
| "loss": 4.2891, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.26490066225165565, | |
| "grad_norm": 0.09952688217163086, | |
| "learning_rate": 0.0009969068351136706, | |
| "loss": 4.2913, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2913907284768212, | |
| "grad_norm": 0.1054968312382698, | |
| "learning_rate": 0.0009956012654497074, | |
| "loss": 4.2794, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.31788079470198677, | |
| "grad_norm": 0.11566981673240662, | |
| "learning_rate": 0.000994067603349975, | |
| "loss": 4.3372, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.3443708609271523, | |
| "grad_norm": 0.11426500976085663, | |
| "learning_rate": 0.0009923065546575544, | |
| "loss": 4.2932, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.3708609271523179, | |
| "grad_norm": 0.1628023236989975, | |
| "learning_rate": 0.000990318929866513, | |
| "loss": 4.3015, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.3973509933774834, | |
| "grad_norm": 0.1205693706870079, | |
| "learning_rate": 0.00098810564374889, | |
| "loss": 4.2702, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.423841059602649, | |
| "grad_norm": 0.10916964709758759, | |
| "learning_rate": 0.0009856677149336858, | |
| "loss": 4.2937, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.4503311258278146, | |
| "grad_norm": 0.1388256847858429, | |
| "learning_rate": 0.0009830062654380548, | |
| "loss": 4.2742, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.4768211920529801, | |
| "grad_norm": 0.17313151061534882, | |
| "learning_rate": 0.0009801225201509158, | |
| "loss": 4.2593, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.5033112582781457, | |
| "grad_norm": 0.13390372693538666, | |
| "learning_rate": 0.0009770178062692165, | |
| "loss": 4.2749, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.5298013245033113, | |
| "grad_norm": 0.1548251509666443, | |
| "learning_rate": 0.000973693552687112, | |
| "loss": 4.2509, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5562913907284768, | |
| "grad_norm": 0.13545338809490204, | |
| "learning_rate": 0.0009701512893383407, | |
| "loss": 4.308, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.5827814569536424, | |
| "grad_norm": 0.12648995220661163, | |
| "learning_rate": 0.0009663926464920958, | |
| "loss": 4.2801, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.609271523178808, | |
| "grad_norm": 0.13534340262413025, | |
| "learning_rate": 0.0009624193540027217, | |
| "loss": 4.2898, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.6357615894039735, | |
| "grad_norm": 0.12199301272630692, | |
| "learning_rate": 0.0009582332405135758, | |
| "loss": 4.2394, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.6622516556291391, | |
| "grad_norm": 0.12553353607654572, | |
| "learning_rate": 0.0009538362326154249, | |
| "loss": 4.2391, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6887417218543046, | |
| "grad_norm": 0.12340961396694183, | |
| "learning_rate": 0.0009492303539597636, | |
| "loss": 4.2676, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 0.7152317880794702, | |
| "grad_norm": 0.10325583815574646, | |
| "learning_rate": 0.0009444177243274617, | |
| "loss": 4.2635, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 0.7417218543046358, | |
| "grad_norm": 0.13559825718402863, | |
| "learning_rate": 0.0009394005586531688, | |
| "loss": 4.2688, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 0.7682119205298014, | |
| "grad_norm": 0.10041986405849457, | |
| "learning_rate": 0.0009341811660059271, | |
| "loss": 4.2227, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 0.7947019867549668, | |
| "grad_norm": 0.1226314976811409, | |
| "learning_rate": 0.0009287619485264596, | |
| "loss": 4.2819, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8211920529801324, | |
| "grad_norm": 0.1134864017367363, | |
| "learning_rate": 0.0009231454003216239, | |
| "loss": 4.2315, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 0.847682119205298, | |
| "grad_norm": 0.1089298278093338, | |
| "learning_rate": 0.0009173341063165405, | |
| "loss": 4.2791, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 0.8741721854304636, | |
| "grad_norm": 0.1100950837135315, | |
| "learning_rate": 0.0009113307410649221, | |
| "loss": 4.2643, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 0.9006622516556292, | |
| "grad_norm": 0.11873114854097366, | |
| "learning_rate": 0.000905138067518154, | |
| "loss": 4.2508, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 0.9271523178807947, | |
| "grad_norm": 0.10352523624897003, | |
| "learning_rate": 0.0008987589357536914, | |
| "loss": 4.277, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9536423841059603, | |
| "grad_norm": 0.0963631346821785, | |
| "learning_rate": 0.0008921962816633561, | |
| "loss": 4.2486, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 0.9801324503311258, | |
| "grad_norm": 0.12682202458381653, | |
| "learning_rate": 0.000885453125602141, | |
| "loss": 4.3274, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.11091629800496128, | |
| "eval_cap_loss": 1.2195911861413362, | |
| "eval_con_loss": 1.8316854106274663, | |
| "eval_loss": 3.0512766001240306, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.11091629800496128, | |
| "eval_cap_loss": 1.2195911861413362, | |
| "eval_con_loss": 1.8316854106274663, | |
| "eval_loss": 3.0512766001240306, | |
| "eval_runtime": 242.8774, | |
| "eval_samples_per_second": 19.882, | |
| "eval_steps_per_second": 2.487, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 1.0066225165562914, | |
| "grad_norm": 0.16604916751384735, | |
| "learning_rate": 0.0008785325709981404, | |
| "loss": 4.226, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 1.033112582781457, | |
| "grad_norm": 0.1470925211906433, | |
| "learning_rate": 0.0008714378029242477, | |
| "loss": 4.2582, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 1.0596026490066226, | |
| "grad_norm": 0.09578409790992737, | |
| "learning_rate": 0.0008641720866322773, | |
| "loss": 4.2641, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.086092715231788, | |
| "grad_norm": 0.13495096564292908, | |
| "learning_rate": 0.0008567387660501852, | |
| "loss": 4.2752, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 1.1125827814569536, | |
| "grad_norm": 0.18220090866088867, | |
| "learning_rate": 0.000849141262243081, | |
| "loss": 4.2859, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 1.1390728476821192, | |
| "grad_norm": 0.10595650970935822, | |
| "learning_rate": 0.0008413830718387375, | |
| "loss": 4.2599, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 1.1655629139072847, | |
| "grad_norm": 0.2255912721157074, | |
| "learning_rate": 0.0008334677654183254, | |
| "loss": 4.2598, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 1.1920529801324504, | |
| "grad_norm": 0.15248772501945496, | |
| "learning_rate": 0.0008253989858731106, | |
| "loss": 4.2828, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.218543046357616, | |
| "grad_norm": 0.10478087514638901, | |
| "learning_rate": 0.0008171804467278729, | |
| "loss": 4.2615, | |
| "step": 736 | |
| }, | |
| { | |
| "epoch": 1.2450331125827814, | |
| "grad_norm": 0.13858726620674133, | |
| "learning_rate": 0.0008088159304318166, | |
| "loss": 4.2932, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 1.271523178807947, | |
| "grad_norm": 0.13052597641944885, | |
| "learning_rate": 0.0008003092866177592, | |
| "loss": 4.2778, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 1.2980132450331126, | |
| "grad_norm": 0.1585925668478012, | |
| "learning_rate": 0.0007916644303304013, | |
| "loss": 4.2626, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 1.3245033112582782, | |
| "grad_norm": 0.15588369965553284, | |
| "learning_rate": 0.0007828853402244896, | |
| "loss": 4.3056, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3509933774834437, | |
| "grad_norm": 0.09688248485326767, | |
| "learning_rate": 0.0007739760567337073, | |
| "loss": 4.2525, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 1.3774834437086092, | |
| "grad_norm": 0.12090156972408295, | |
| "learning_rate": 0.0007649406802111283, | |
| "loss": 4.223, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 1.403973509933775, | |
| "grad_norm": 0.11892067641019821, | |
| "learning_rate": 0.0007557833690420974, | |
| "loss": 4.248, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 1.4304635761589404, | |
| "grad_norm": 0.1257970631122589, | |
| "learning_rate": 0.0007465083377304009, | |
| "loss": 4.2604, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 1.4569536423841059, | |
| "grad_norm": 0.1225869208574295, | |
| "learning_rate": 0.0007371198549586091, | |
| "loss": 4.2583, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.4834437086092715, | |
| "grad_norm": 0.12114247679710388, | |
| "learning_rate": 0.000727622241623485, | |
| "loss": 4.2293, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 1.5099337748344372, | |
| "grad_norm": 0.14327970147132874, | |
| "learning_rate": 0.0007180198688473614, | |
| "loss": 4.3047, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 1.5364238410596025, | |
| "grad_norm": 0.12314064800739288, | |
| "learning_rate": 0.0007083171559664032, | |
| "loss": 4.2156, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 1.5629139072847682, | |
| "grad_norm": 0.10841790586709976, | |
| "learning_rate": 0.0006985185684966791, | |
| "loss": 4.2602, | |
| "step": 944 | |
| }, | |
| { | |
| "epoch": 1.589403973509934, | |
| "grad_norm": 0.13060128688812256, | |
| "learning_rate": 0.0006886286160789805, | |
| "loss": 4.2554, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.6158940397350994, | |
| "grad_norm": 0.10328269004821777, | |
| "learning_rate": 0.0006786518504033333, | |
| "loss": 4.2572, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 1.6423841059602649, | |
| "grad_norm": 0.15084649622440338, | |
| "learning_rate": 0.0006685928631141552, | |
| "loss": 4.278, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 1.6688741721854305, | |
| "grad_norm": 0.1322745680809021, | |
| "learning_rate": 0.0006584562836970271, | |
| "loss": 4.2747, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 1.695364238410596, | |
| "grad_norm": 0.09931815415620804, | |
| "learning_rate": 0.0006482467773480468, | |
| "loss": 4.2314, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 1.7218543046357615, | |
| "grad_norm": 0.12023808807134628, | |
| "learning_rate": 0.0006379690428267482, | |
| "loss": 4.2832, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.7483443708609272, | |
| "grad_norm": 0.12222672253847122, | |
| "learning_rate": 0.000627627810293574, | |
| "loss": 4.1856, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 1.7748344370860927, | |
| "grad_norm": 0.14069874584674835, | |
| "learning_rate": 0.0006172278391328957, | |
| "loss": 4.2394, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 1.8013245033112582, | |
| "grad_norm": 0.12728360295295715, | |
| "learning_rate": 0.0006067739157625848, | |
| "loss": 4.3155, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 1.8278145695364238, | |
| "grad_norm": 0.15248538553714752, | |
| "learning_rate": 0.0005962708514311411, | |
| "loss": 4.2606, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 1.8543046357615895, | |
| "grad_norm": 0.08556041121482849, | |
| "learning_rate": 0.0005857234800033936, | |
| "loss": 4.2379, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.8807947019867548, | |
| "grad_norm": 0.10861990600824356, | |
| "learning_rate": 0.0005751366557357933, | |
| "loss": 4.2761, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 1.9072847682119205, | |
| "grad_norm": 0.1670146882534027, | |
| "learning_rate": 0.0005645152510423204, | |
| "loss": 4.2325, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 1.9337748344370862, | |
| "grad_norm": 0.14551334083080292, | |
| "learning_rate": 0.0005538641542520343, | |
| "loss": 4.2249, | |
| "step": 1168 | |
| }, | |
| { | |
| "epoch": 1.9602649006622517, | |
| "grad_norm": 0.12024614959955215, | |
| "learning_rate": 0.0005431882673592999, | |
| "loss": 4.2265, | |
| "step": 1184 | |
| }, | |
| { | |
| "epoch": 1.9867549668874172, | |
| "grad_norm": 0.15291069447994232, | |
| "learning_rate": 0.0005324925037677243, | |
| "loss": 4.3172, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.11370215395599652, | |
| "eval_cap_loss": 1.2129609353889692, | |
| "eval_con_loss": 1.808429909265594, | |
| "eval_loss": 3.0213908505755542, | |
| "step": 1208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.11370215395599652, | |
| "eval_cap_loss": 1.2129609353889692, | |
| "eval_con_loss": 1.808429909265594, | |
| "eval_loss": 3.0213908505755542, | |
| "eval_runtime": 241.5496, | |
| "eval_samples_per_second": 19.992, | |
| "eval_steps_per_second": 2.501, | |
| "step": 1208 | |
| } | |
| ], | |
| "logging_steps": 16, | |
| "max_steps": 2416, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null, | |
| "tau_value": 4.5962 | |
| } | |