| { | |
| "best_metric": 0.8237398652434835, | |
| "best_model_checkpoint": "output/eurobert_simce_EuroBERT-EuroBERT-210m_32_bs_1_e/checkpoint-4500", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 7813, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02559836170485089, | |
| "grad_norm": 21741072.0, | |
| "learning_rate": 1.2787723785166241e-05, | |
| "loss": 8.8816, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05119672340970178, | |
| "grad_norm": 11201852.0, | |
| "learning_rate": 2.5575447570332482e-05, | |
| "loss": 5.1404, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06399590426212723, | |
| "eval_loss": 6.53036642074585, | |
| "eval_runtime": 185.969, | |
| "eval_samples_per_second": 35.538, | |
| "eval_sequential_score": 0.785466694337272, | |
| "eval_steps_per_second": 0.28, | |
| "eval_sts-dev-128_pearson_cosine": 0.763999536812149, | |
| "eval_sts-dev-128_spearman_cosine": 0.7711764958651725, | |
| "eval_sts-dev-256_pearson_cosine": 0.7720362664289375, | |
| "eval_sts-dev-256_spearman_cosine": 0.7765855658716928, | |
| "eval_sts-dev-512_pearson_cosine": 0.7788408953704133, | |
| "eval_sts-dev-512_spearman_cosine": 0.7817962751922598, | |
| "eval_sts-dev-64_pearson_cosine": 0.7530255624130506, | |
| "eval_sts-dev-64_spearman_cosine": 0.7635133266587258, | |
| "eval_sts-dev-768_pearson_cosine": 0.7833066834602731, | |
| "eval_sts-dev-768_spearman_cosine": 0.785466694337272, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07679508511455267, | |
| "grad_norm": 97525040.0, | |
| "learning_rate": 3.8363171355498725e-05, | |
| "loss": 4.7789, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.10239344681940356, | |
| "grad_norm": 16968314.0, | |
| "learning_rate": 4.987199544872707e-05, | |
| "loss": 4.6845, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.12799180852425446, | |
| "grad_norm": 8517995.0, | |
| "learning_rate": 4.844972265680558e-05, | |
| "loss": 4.6628, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12799180852425446, | |
| "eval_loss": 6.329844951629639, | |
| "eval_runtime": 443.2691, | |
| "eval_samples_per_second": 14.91, | |
| "eval_sequential_score": 0.7900474282041248, | |
| "eval_steps_per_second": 0.117, | |
| "eval_sts-dev-128_pearson_cosine": 0.7796253891599777, | |
| "eval_sts-dev-128_spearman_cosine": 0.7828728829847589, | |
| "eval_sts-dev-256_pearson_cosine": 0.7844919892349245, | |
| "eval_sts-dev-256_spearman_cosine": 0.7854868395314357, | |
| "eval_sts-dev-512_pearson_cosine": 0.7895438861374281, | |
| "eval_sts-dev-512_spearman_cosine": 0.7887806711839089, | |
| "eval_sts-dev-64_pearson_cosine": 0.7711864442090859, | |
| "eval_sts-dev-64_spearman_cosine": 0.7756594477323392, | |
| "eval_sts-dev-768_pearson_cosine": 0.7912711696792909, | |
| "eval_sts-dev-768_spearman_cosine": 0.7900474282041248, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15359017022910534, | |
| "grad_norm": 9523596.0, | |
| "learning_rate": 4.702744986488409e-05, | |
| "loss": 4.2947, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.17918853193395623, | |
| "grad_norm": 12880760.0, | |
| "learning_rate": 4.56051770729626e-05, | |
| "loss": 4.0669, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.19198771278638166, | |
| "eval_loss": 6.11918830871582, | |
| "eval_runtime": 442.8341, | |
| "eval_samples_per_second": 14.924, | |
| "eval_sequential_score": 0.776235475618058, | |
| "eval_steps_per_second": 0.117, | |
| "eval_sts-dev-128_pearson_cosine": 0.7556070600058153, | |
| "eval_sts-dev-128_spearman_cosine": 0.7650883993782764, | |
| "eval_sts-dev-256_pearson_cosine": 0.7615982531224779, | |
| "eval_sts-dev-256_spearman_cosine": 0.7688372980643342, | |
| "eval_sts-dev-512_pearson_cosine": 0.7658162533685451, | |
| "eval_sts-dev-512_spearman_cosine": 0.7717076619559236, | |
| "eval_sts-dev-64_pearson_cosine": 0.7412214467196323, | |
| "eval_sts-dev-64_spearman_cosine": 0.7546465937668674, | |
| "eval_sts-dev-768_pearson_cosine": 0.7709585212331336, | |
| "eval_sts-dev-768_spearman_cosine": 0.776235475618058, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.20478689363880712, | |
| "grad_norm": 5121008.5, | |
| "learning_rate": 4.4182904281041105e-05, | |
| "loss": 3.7798, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.230385255343658, | |
| "grad_norm": 8532118.0, | |
| "learning_rate": 4.276063148911961e-05, | |
| "loss": 3.6295, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.2559836170485089, | |
| "grad_norm": 5177615.0, | |
| "learning_rate": 4.133835869719813e-05, | |
| "loss": 3.4326, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2559836170485089, | |
| "eval_loss": 5.5250773429870605, | |
| "eval_runtime": 129.0233, | |
| "eval_samples_per_second": 51.223, | |
| "eval_sequential_score": 0.7967876172773333, | |
| "eval_steps_per_second": 0.403, | |
| "eval_sts-dev-128_pearson_cosine": 0.784502590249166, | |
| "eval_sts-dev-128_spearman_cosine": 0.7904922950886314, | |
| "eval_sts-dev-256_pearson_cosine": 0.7879728763645151, | |
| "eval_sts-dev-256_spearman_cosine": 0.7926398088881677, | |
| "eval_sts-dev-512_pearson_cosine": 0.79034320657993, | |
| "eval_sts-dev-512_spearman_cosine": 0.7941246347219933, | |
| "eval_sts-dev-64_pearson_cosine": 0.7732918356795019, | |
| "eval_sts-dev-64_spearman_cosine": 0.7822318891319547, | |
| "eval_sts-dev-768_pearson_cosine": 0.7932888301272122, | |
| "eval_sts-dev-768_spearman_cosine": 0.7967876172773333, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2815819787533598, | |
| "grad_norm": 3656165.0, | |
| "learning_rate": 3.9916085905276635e-05, | |
| "loss": 3.5024, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.3071803404582107, | |
| "grad_norm": 4165362.5, | |
| "learning_rate": 3.849381311335514e-05, | |
| "loss": 3.2039, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.3199795213106361, | |
| "eval_loss": 5.417263507843018, | |
| "eval_runtime": 190.9081, | |
| "eval_samples_per_second": 34.619, | |
| "eval_sequential_score": 0.7985143359688982, | |
| "eval_steps_per_second": 0.272, | |
| "eval_sts-dev-128_pearson_cosine": 0.7806598252859374, | |
| "eval_sts-dev-128_spearman_cosine": 0.7903599418037432, | |
| "eval_sts-dev-256_pearson_cosine": 0.7884086400168866, | |
| "eval_sts-dev-256_spearman_cosine": 0.7946240645451771, | |
| "eval_sts-dev-512_pearson_cosine": 0.7915267570052114, | |
| "eval_sts-dev-512_spearman_cosine": 0.7957194058029897, | |
| "eval_sts-dev-64_pearson_cosine": 0.7654385410326767, | |
| "eval_sts-dev-64_spearman_cosine": 0.7805870680928543, | |
| "eval_sts-dev-768_pearson_cosine": 0.7950747299618343, | |
| "eval_sts-dev-768_spearman_cosine": 0.7985143359688982, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.33277870216306155, | |
| "grad_norm": 3111441.75, | |
| "learning_rate": 3.707154032143365e-05, | |
| "loss": 3.1517, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.35837706386791246, | |
| "grad_norm": 4315246.0, | |
| "learning_rate": 3.5649267529512165e-05, | |
| "loss": 3.0409, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.3839754255727633, | |
| "grad_norm": 4406995.0, | |
| "learning_rate": 3.422699473759067e-05, | |
| "loss": 2.9611, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3839754255727633, | |
| "eval_loss": 5.039449691772461, | |
| "eval_runtime": 186.7714, | |
| "eval_samples_per_second": 35.385, | |
| "eval_sequential_score": 0.7922583016547031, | |
| "eval_steps_per_second": 0.278, | |
| "eval_sts-dev-128_pearson_cosine": 0.782435760922896, | |
| "eval_sts-dev-128_spearman_cosine": 0.7847986500402634, | |
| "eval_sts-dev-256_pearson_cosine": 0.785603571035357, | |
| "eval_sts-dev-256_spearman_cosine": 0.7871271821843686, | |
| "eval_sts-dev-512_pearson_cosine": 0.7882585155923556, | |
| "eval_sts-dev-512_spearman_cosine": 0.7893899811664858, | |
| "eval_sts-dev-64_pearson_cosine": 0.7731641525494573, | |
| "eval_sts-dev-64_spearman_cosine": 0.7789396674425884, | |
| "eval_sts-dev-768_pearson_cosine": 0.7914921308293312, | |
| "eval_sts-dev-768_spearman_cosine": 0.7922583016547031, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.40957378727761423, | |
| "grad_norm": 4295917.5, | |
| "learning_rate": 3.280472194566918e-05, | |
| "loss": 2.8913, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.43517214898246515, | |
| "grad_norm": 3956720.0, | |
| "learning_rate": 3.138244915374769e-05, | |
| "loss": 2.6737, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.4479713298348906, | |
| "eval_loss": 4.8449788093566895, | |
| "eval_runtime": 381.2673, | |
| "eval_samples_per_second": 17.334, | |
| "eval_sequential_score": 0.8124001874236463, | |
| "eval_steps_per_second": 0.136, | |
| "eval_sts-dev-128_pearson_cosine": 0.7982139278806232, | |
| "eval_sts-dev-128_spearman_cosine": 0.8076074750236868, | |
| "eval_sts-dev-256_pearson_cosine": 0.7999244325189871, | |
| "eval_sts-dev-256_spearman_cosine": 0.8075358060747592, | |
| "eval_sts-dev-512_pearson_cosine": 0.8054012343851129, | |
| "eval_sts-dev-512_spearman_cosine": 0.8110777459628828, | |
| "eval_sts-dev-64_pearson_cosine": 0.7826221833277923, | |
| "eval_sts-dev-64_spearman_cosine": 0.7968397814525646, | |
| "eval_sts-dev-768_pearson_cosine": 0.8074395668796857, | |
| "eval_sts-dev-768_spearman_cosine": 0.8124001874236463, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.460770510687316, | |
| "grad_norm": 3219552.5, | |
| "learning_rate": 2.99601763618262e-05, | |
| "loss": 2.6488, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.4863688723921669, | |
| "grad_norm": 3358243.75, | |
| "learning_rate": 2.853790356990471e-05, | |
| "loss": 2.6208, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.5119672340970178, | |
| "grad_norm": 4434459.5, | |
| "learning_rate": 2.7115630777983218e-05, | |
| "loss": 2.4823, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5119672340970178, | |
| "eval_loss": 4.5710768699646, | |
| "eval_runtime": 216.5327, | |
| "eval_samples_per_second": 30.522, | |
| "eval_sequential_score": 0.811115353774689, | |
| "eval_steps_per_second": 0.24, | |
| "eval_sts-dev-128_pearson_cosine": 0.7996500780619267, | |
| "eval_sts-dev-128_spearman_cosine": 0.8075174124731305, | |
| "eval_sts-dev-256_pearson_cosine": 0.8022455109638521, | |
| "eval_sts-dev-256_spearman_cosine": 0.8081526580763048, | |
| "eval_sts-dev-512_pearson_cosine": 0.8053149776225357, | |
| "eval_sts-dev-512_spearman_cosine": 0.8101542245323032, | |
| "eval_sts-dev-64_pearson_cosine": 0.7897403224233863, | |
| "eval_sts-dev-64_spearman_cosine": 0.8014649100236256, | |
| "eval_sts-dev-768_pearson_cosine": 0.8071703574474214, | |
| "eval_sts-dev-768_spearman_cosine": 0.811115353774689, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5375655958018687, | |
| "grad_norm": 3851683.75, | |
| "learning_rate": 2.569335798606173e-05, | |
| "loss": 2.5081, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.5631639575067195, | |
| "grad_norm": 3974658.0, | |
| "learning_rate": 2.4271085194140237e-05, | |
| "loss": 2.3827, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.575963138359145, | |
| "eval_loss": 4.527626037597656, | |
| "eval_runtime": 247.0855, | |
| "eval_samples_per_second": 26.748, | |
| "eval_sequential_score": 0.8237398652434835, | |
| "eval_steps_per_second": 0.21, | |
| "eval_sts-dev-128_pearson_cosine": 0.8088977658009835, | |
| "eval_sts-dev-128_spearman_cosine": 0.8200038307453663, | |
| "eval_sts-dev-256_pearson_cosine": 0.8120809058779974, | |
| "eval_sts-dev-256_spearman_cosine": 0.8205438030370273, | |
| "eval_sts-dev-512_pearson_cosine": 0.8157264805096027, | |
| "eval_sts-dev-512_spearman_cosine": 0.8226500864435473, | |
| "eval_sts-dev-64_pearson_cosine": 0.7964069841493276, | |
| "eval_sts-dev-64_spearman_cosine": 0.8117000425044992, | |
| "eval_sts-dev-768_pearson_cosine": 0.8179809027318157, | |
| "eval_sts-dev-768_spearman_cosine": 0.8237398652434835, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5887623192115704, | |
| "grad_norm": 2969929.75, | |
| "learning_rate": 2.284881240221875e-05, | |
| "loss": 2.2867, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.6143606809164214, | |
| "grad_norm": 3068778.0, | |
| "learning_rate": 2.1426539610297256e-05, | |
| "loss": 2.2608, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.6399590426212722, | |
| "grad_norm": 5789758.5, | |
| "learning_rate": 2.0004266818375767e-05, | |
| "loss": 2.6285, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6399590426212722, | |
| "eval_loss": 2.69280743598938, | |
| "eval_runtime": 587.2079, | |
| "eval_samples_per_second": 11.255, | |
| "eval_sequential_score": 0.812358680832537, | |
| "eval_steps_per_second": 0.089, | |
| "eval_sts-dev-128_pearson_cosine": 0.805215933596938, | |
| "eval_sts-dev-128_spearman_cosine": 0.8087200580616569, | |
| "eval_sts-dev-256_pearson_cosine": 0.8085076472836253, | |
| "eval_sts-dev-256_spearman_cosine": 0.8099002132418758, | |
| "eval_sts-dev-512_pearson_cosine": 0.8104747450142471, | |
| "eval_sts-dev-512_spearman_cosine": 0.8112676803940946, | |
| "eval_sts-dev-64_pearson_cosine": 0.7956052250413164, | |
| "eval_sts-dev-64_spearman_cosine": 0.8022672223914163, | |
| "eval_sts-dev-768_pearson_cosine": 0.8122169140710348, | |
| "eval_sts-dev-768_spearman_cosine": 0.812358680832537, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6655574043261231, | |
| "grad_norm": 7265908.5, | |
| "learning_rate": 1.8581994026454275e-05, | |
| "loss": 3.2569, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.6911557660309741, | |
| "grad_norm": 5744435.0, | |
| "learning_rate": 1.7159721234532783e-05, | |
| "loss": 2.7108, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.7039549468833994, | |
| "eval_loss": 3.4081127643585205, | |
| "eval_runtime": 113.2957, | |
| "eval_samples_per_second": 58.334, | |
| "eval_sequential_score": 0.8112072214643352, | |
| "eval_steps_per_second": 0.459, | |
| "eval_sts-dev-128_pearson_cosine": 0.8036120610302749, | |
| "eval_sts-dev-128_spearman_cosine": 0.8060065068978162, | |
| "eval_sts-dev-256_pearson_cosine": 0.8076653976665353, | |
| "eval_sts-dev-256_spearman_cosine": 0.8079945036667597, | |
| "eval_sts-dev-512_pearson_cosine": 0.8103438010262101, | |
| "eval_sts-dev-512_spearman_cosine": 0.8099837098639602, | |
| "eval_sts-dev-64_pearson_cosine": 0.7942652314065892, | |
| "eval_sts-dev-64_spearman_cosine": 0.7993950394097328, | |
| "eval_sts-dev-768_pearson_cosine": 0.8122845848971061, | |
| "eval_sts-dev-768_spearman_cosine": 0.8112072214643352, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7167541277358249, | |
| "grad_norm": 5608753.5, | |
| "learning_rate": 1.5737448442611294e-05, | |
| "loss": 2.2756, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.7423524894406758, | |
| "grad_norm": 4639111.5, | |
| "learning_rate": 1.4315175650689802e-05, | |
| "loss": 1.9964, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.7679508511455266, | |
| "grad_norm": 3252142.5, | |
| "learning_rate": 1.2892902858768313e-05, | |
| "loss": 1.8278, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7679508511455266, | |
| "eval_loss": 3.626107931137085, | |
| "eval_runtime": 402.388, | |
| "eval_samples_per_second": 16.424, | |
| "eval_sequential_score": 0.8115702497698598, | |
| "eval_steps_per_second": 0.129, | |
| "eval_sts-dev-128_pearson_cosine": 0.8045525818531101, | |
| "eval_sts-dev-128_spearman_cosine": 0.8070988970713203, | |
| "eval_sts-dev-256_pearson_cosine": 0.808009240681977, | |
| "eval_sts-dev-256_spearman_cosine": 0.8087801893710216, | |
| "eval_sts-dev-512_pearson_cosine": 0.8100529203161405, | |
| "eval_sts-dev-512_spearman_cosine": 0.8101431817309978, | |
| "eval_sts-dev-64_pearson_cosine": 0.7958099227865137, | |
| "eval_sts-dev-64_spearman_cosine": 0.8012609269210379, | |
| "eval_sts-dev-768_pearson_cosine": 0.8121800215134509, | |
| "eval_sts-dev-768_spearman_cosine": 0.8115702497698598, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7935492128503776, | |
| "grad_norm": 7875703.0, | |
| "learning_rate": 1.147063006684682e-05, | |
| "loss": 1.7105, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.8191475745552285, | |
| "grad_norm": 4817139.5, | |
| "learning_rate": 1.0048357274925332e-05, | |
| "loss": 1.5719, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.831946755407654, | |
| "eval_loss": 3.7825615406036377, | |
| "eval_runtime": 260.9688, | |
| "eval_samples_per_second": 25.325, | |
| "eval_sequential_score": 0.809746911527668, | |
| "eval_steps_per_second": 0.199, | |
| "eval_sts-dev-128_pearson_cosine": 0.8019992344978342, | |
| "eval_sts-dev-128_spearman_cosine": 0.8040026735198376, | |
| "eval_sts-dev-256_pearson_cosine": 0.8069582374847454, | |
| "eval_sts-dev-256_spearman_cosine": 0.8072454625835676, | |
| "eval_sts-dev-512_pearson_cosine": 0.8088522922289604, | |
| "eval_sts-dev-512_spearman_cosine": 0.8085037600756491, | |
| "eval_sts-dev-64_pearson_cosine": 0.7917208895813888, | |
| "eval_sts-dev-64_spearman_cosine": 0.7966340332099681, | |
| "eval_sts-dev-768_pearson_cosine": 0.8110657541385573, | |
| "eval_sts-dev-768_spearman_cosine": 0.809746911527668, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.8447459362600793, | |
| "grad_norm": 3659550.25, | |
| "learning_rate": 8.626084483003841e-06, | |
| "loss": 1.4569, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.8703442979649303, | |
| "grad_norm": 5161893.0, | |
| "learning_rate": 7.20381169108235e-06, | |
| "loss": 1.3572, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.8959426596697811, | |
| "grad_norm": 3728513.5, | |
| "learning_rate": 5.781538899160859e-06, | |
| "loss": 1.2607, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8959426596697811, | |
| "eval_loss": 3.732253074645996, | |
| "eval_runtime": 343.8, | |
| "eval_samples_per_second": 19.223, | |
| "eval_sequential_score": 0.8113987863085653, | |
| "eval_steps_per_second": 0.151, | |
| "eval_sts-dev-128_pearson_cosine": 0.8052203471354245, | |
| "eval_sts-dev-128_spearman_cosine": 0.8069675064136675, | |
| "eval_sts-dev-256_pearson_cosine": 0.8090504276321419, | |
| "eval_sts-dev-256_spearman_cosine": 0.8092690932878782, | |
| "eval_sts-dev-512_pearson_cosine": 0.8106848356090132, | |
| "eval_sts-dev-512_spearman_cosine": 0.8102076026050016, | |
| "eval_sts-dev-64_pearson_cosine": 0.7959836620564662, | |
| "eval_sts-dev-64_spearman_cosine": 0.8004825005334193, | |
| "eval_sts-dev-768_pearson_cosine": 0.8127183729174601, | |
| "eval_sts-dev-768_spearman_cosine": 0.8113987863085653, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.921541021374632, | |
| "grad_norm": 3384944.0, | |
| "learning_rate": 4.359266107239369e-06, | |
| "loss": 1.1676, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.9471393830794829, | |
| "grad_norm": 3170591.5, | |
| "learning_rate": 2.936993315317878e-06, | |
| "loss": 1.1663, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.9599385639319084, | |
| "eval_loss": 3.83072829246521, | |
| "eval_runtime": 240.3095, | |
| "eval_samples_per_second": 27.502, | |
| "eval_sequential_score": 0.8100586279907306, | |
| "eval_steps_per_second": 0.216, | |
| "eval_sts-dev-128_pearson_cosine": 0.8028710019029521, | |
| "eval_sts-dev-128_spearman_cosine": 0.8054855987917489, | |
| "eval_sts-dev-256_pearson_cosine": 0.8076510620939634, | |
| "eval_sts-dev-256_spearman_cosine": 0.8080588277305082, | |
| "eval_sts-dev-512_pearson_cosine": 0.8092891955563192, | |
| "eval_sts-dev-512_spearman_cosine": 0.8087644228771842, | |
| "eval_sts-dev-64_pearson_cosine": 0.7923252906438638, | |
| "eval_sts-dev-64_spearman_cosine": 0.7975941111911333, | |
| "eval_sts-dev-768_pearson_cosine": 0.8111988062913815, | |
| "eval_sts-dev-768_spearman_cosine": 0.8100586279907306, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.9727377447843338, | |
| "grad_norm": 3539899.5, | |
| "learning_rate": 1.5147205233963876e-06, | |
| "loss": 1.1079, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.9983361064891847, | |
| "grad_norm": 4290327.5, | |
| "learning_rate": 9.24477314748969e-08, | |
| "loss": 1.0827, | |
| "step": 7800 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 7813, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |