| { | |
| "best_metric": 1.0273813009262085, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/dan-Latn/checkpoint-98500", | |
| "epoch": 10.286131996658312, | |
| "eval_steps": 500, | |
| "global_step": 98500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.052213868003341685, | |
| "grad_norm": 5.384862422943115, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.4559, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.052213868003341685, | |
| "eval_accuracy": 0.7296120738961336, | |
| "eval_loss": 1.6221407651901245, | |
| "eval_runtime": 620.8615, | |
| "eval_samples_per_second": 102.185, | |
| "eval_steps_per_second": 3.194, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10442773600668337, | |
| "grad_norm": 4.473601818084717, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.4355, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10442773600668337, | |
| "eval_accuracy": 0.7314203520980975, | |
| "eval_loss": 1.563644289970398, | |
| "eval_runtime": 631.0902, | |
| "eval_samples_per_second": 100.529, | |
| "eval_steps_per_second": 3.142, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "grad_norm": 4.604930877685547, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.4326, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "eval_accuracy": 0.732858138551035, | |
| "eval_loss": 1.538730502128601, | |
| "eval_runtime": 627.1459, | |
| "eval_samples_per_second": 101.161, | |
| "eval_steps_per_second": 3.162, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.20885547201336674, | |
| "grad_norm": 5.0758867263793945, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.4082, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.20885547201336674, | |
| "eval_accuracy": 0.7337186853777712, | |
| "eval_loss": 1.514768123626709, | |
| "eval_runtime": 623.7693, | |
| "eval_samples_per_second": 101.709, | |
| "eval_steps_per_second": 3.179, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.26106934001670845, | |
| "grad_norm": 4.955592155456543, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.4066, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.26106934001670845, | |
| "eval_accuracy": 0.7373838924395351, | |
| "eval_loss": 1.4860447645187378, | |
| "eval_runtime": 626.6763, | |
| "eval_samples_per_second": 101.237, | |
| "eval_steps_per_second": 3.164, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "grad_norm": 4.4695611000061035, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.3965, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "eval_accuracy": 0.7386060798818527, | |
| "eval_loss": 1.4720885753631592, | |
| "eval_runtime": 626.8807, | |
| "eval_samples_per_second": 101.204, | |
| "eval_steps_per_second": 3.163, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3654970760233918, | |
| "grad_norm": 4.595192909240723, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.3674, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3654970760233918, | |
| "eval_accuracy": 0.7389515252696706, | |
| "eval_loss": 1.466160774230957, | |
| "eval_runtime": 626.2854, | |
| "eval_samples_per_second": 101.3, | |
| "eval_steps_per_second": 3.166, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.4177109440267335, | |
| "grad_norm": 4.489499092102051, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.3789, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4177109440267335, | |
| "eval_accuracy": 0.7408971720895672, | |
| "eval_loss": 1.4427233934402466, | |
| "eval_runtime": 628.8309, | |
| "eval_samples_per_second": 100.89, | |
| "eval_steps_per_second": 3.153, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "grad_norm": 4.887514114379883, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.3677, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "eval_accuracy": 0.7418176415206207, | |
| "eval_loss": NaN, | |
| "eval_runtime": 626.3841, | |
| "eval_samples_per_second": 101.285, | |
| "eval_steps_per_second": 3.166, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5221386800334169, | |
| "grad_norm": 4.255401611328125, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.3591, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5221386800334169, | |
| "eval_accuracy": 0.7420801988701987, | |
| "eval_loss": 1.4378492832183838, | |
| "eval_runtime": 628.3584, | |
| "eval_samples_per_second": 100.966, | |
| "eval_steps_per_second": 3.156, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5743525480367586, | |
| "grad_norm": 3.9552218914031982, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.3528, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5743525480367586, | |
| "eval_accuracy": 0.7430785963224286, | |
| "eval_loss": 1.4063905477523804, | |
| "eval_runtime": 630.6125, | |
| "eval_samples_per_second": 100.605, | |
| "eval_steps_per_second": 3.145, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6265664160401002, | |
| "grad_norm": 4.043545246124268, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.3467, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6265664160401002, | |
| "eval_accuracy": 0.7439164901309249, | |
| "eval_loss": 1.4179232120513916, | |
| "eval_runtime": 627.0519, | |
| "eval_samples_per_second": 101.177, | |
| "eval_steps_per_second": 3.162, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6787802840434419, | |
| "grad_norm": 65.49678802490234, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.3578, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6787802840434419, | |
| "eval_accuracy": 0.7436562018580148, | |
| "eval_loss": 1.4101091623306274, | |
| "eval_runtime": 631.9563, | |
| "eval_samples_per_second": 100.391, | |
| "eval_steps_per_second": 3.138, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7309941520467836, | |
| "grad_norm": 3.9676353931427, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 1.3377, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7309941520467836, | |
| "eval_accuracy": 0.7452273902409787, | |
| "eval_loss": 1.3957942724227905, | |
| "eval_runtime": 625.612, | |
| "eval_samples_per_second": 101.41, | |
| "eval_steps_per_second": 3.17, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7832080200501254, | |
| "grad_norm": 3.6173226833343506, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 1.3303, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7832080200501254, | |
| "eval_accuracy": 0.7475488951341268, | |
| "eval_loss": 1.3813687562942505, | |
| "eval_runtime": 627.0109, | |
| "eval_samples_per_second": 101.183, | |
| "eval_steps_per_second": 3.163, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.835421888053467, | |
| "grad_norm": 3.980341911315918, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.3221, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.835421888053467, | |
| "eval_accuracy": 0.7475162804135862, | |
| "eval_loss": 1.3788014650344849, | |
| "eval_runtime": 625.2826, | |
| "eval_samples_per_second": 101.463, | |
| "eval_steps_per_second": 3.171, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8876357560568087, | |
| "grad_norm": 3.367668390274048, | |
| "learning_rate": 9.15e-05, | |
| "loss": 1.3273, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8876357560568087, | |
| "eval_accuracy": 0.7486955192165756, | |
| "eval_loss": 1.3720810413360596, | |
| "eval_runtime": 626.1965, | |
| "eval_samples_per_second": 101.315, | |
| "eval_steps_per_second": 3.167, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9398496240601504, | |
| "grad_norm": 3.82468843460083, | |
| "learning_rate": 9.1e-05, | |
| "loss": 1.3104, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9398496240601504, | |
| "eval_accuracy": 0.7470319869853385, | |
| "eval_loss": 1.3732112646102905, | |
| "eval_runtime": 630.8001, | |
| "eval_samples_per_second": 100.575, | |
| "eval_steps_per_second": 3.144, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9920634920634921, | |
| "grad_norm": 3.946758270263672, | |
| "learning_rate": 9.05e-05, | |
| "loss": 1.3158, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9920634920634921, | |
| "eval_accuracy": 0.7499583551090239, | |
| "eval_loss": 1.3603472709655762, | |
| "eval_runtime": 652.6391, | |
| "eval_samples_per_second": 97.21, | |
| "eval_steps_per_second": 3.038, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0442773600668338, | |
| "grad_norm": 3.916361093521118, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2953, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0442773600668338, | |
| "eval_accuracy": 0.7500646093431366, | |
| "eval_loss": 1.3458328247070312, | |
| "eval_runtime": 626.4497, | |
| "eval_samples_per_second": 101.274, | |
| "eval_steps_per_second": 3.165, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0964912280701755, | |
| "grad_norm": 3.9930801391601562, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 1.2932, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0964912280701755, | |
| "eval_accuracy": 0.7504904150215282, | |
| "eval_loss": 1.3514128923416138, | |
| "eval_runtime": 630.1314, | |
| "eval_samples_per_second": 100.682, | |
| "eval_steps_per_second": 3.147, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.1487050960735172, | |
| "grad_norm": 4.74911642074585, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 1.2944, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.1487050960735172, | |
| "eval_accuracy": 0.7512302865503326, | |
| "eval_loss": 1.3358726501464844, | |
| "eval_runtime": 631.5659, | |
| "eval_samples_per_second": 100.453, | |
| "eval_steps_per_second": 3.14, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2009189640768587, | |
| "grad_norm": 3.7969741821289062, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 1.2853, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2009189640768587, | |
| "eval_accuracy": 0.7516760587886216, | |
| "eval_loss": 1.3410056829452515, | |
| "eval_runtime": 631.8461, | |
| "eval_samples_per_second": 100.409, | |
| "eval_steps_per_second": 3.138, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2531328320802004, | |
| "grad_norm": 4.402622699737549, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.2904, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.2531328320802004, | |
| "eval_accuracy": 0.7529186511157855, | |
| "eval_loss": 1.3296958208084106, | |
| "eval_runtime": 625.8019, | |
| "eval_samples_per_second": 101.379, | |
| "eval_steps_per_second": 3.169, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.3053467000835421, | |
| "grad_norm": 4.738062858581543, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.2768, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.3053467000835421, | |
| "eval_accuracy": 0.7535759135206747, | |
| "eval_loss": 1.323601484298706, | |
| "eval_runtime": 625.8627, | |
| "eval_samples_per_second": 101.369, | |
| "eval_steps_per_second": 3.168, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.3575605680868839, | |
| "grad_norm": 3.4203574657440186, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.2702, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.3575605680868839, | |
| "eval_accuracy": 0.7543550936371498, | |
| "eval_loss": 1.311880350112915, | |
| "eval_runtime": 632.5695, | |
| "eval_samples_per_second": 100.294, | |
| "eval_steps_per_second": 3.135, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.4097744360902256, | |
| "grad_norm": 4.80487060546875, | |
| "learning_rate": 8.65e-05, | |
| "loss": 1.2686, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.4097744360902256, | |
| "eval_accuracy": 0.7551145942334886, | |
| "eval_loss": 1.3116217851638794, | |
| "eval_runtime": 622.4069, | |
| "eval_samples_per_second": 101.932, | |
| "eval_steps_per_second": 3.186, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.4619883040935673, | |
| "grad_norm": 3.3763253688812256, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.256, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.4619883040935673, | |
| "eval_accuracy": 0.7550032122033714, | |
| "eval_loss": NaN, | |
| "eval_runtime": 623.0788, | |
| "eval_samples_per_second": 101.822, | |
| "eval_steps_per_second": 3.183, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.514202172096909, | |
| "grad_norm": 3.4600539207458496, | |
| "learning_rate": 8.55e-05, | |
| "loss": 1.2689, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.514202172096909, | |
| "eval_accuracy": 0.7562542149486124, | |
| "eval_loss": 1.2966333627700806, | |
| "eval_runtime": 630.2089, | |
| "eval_samples_per_second": 100.67, | |
| "eval_steps_per_second": 3.147, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.5664160401002505, | |
| "grad_norm": 3.6656503677368164, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.2536, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.5664160401002505, | |
| "eval_accuracy": 0.7558193144413351, | |
| "eval_loss": 1.3074419498443604, | |
| "eval_runtime": 633.7511, | |
| "eval_samples_per_second": 100.107, | |
| "eval_steps_per_second": 3.129, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.6186299081035922, | |
| "grad_norm": 3.7582554817199707, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 1.262, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.6186299081035922, | |
| "eval_accuracy": 0.755967622936909, | |
| "eval_loss": 1.2850651741027832, | |
| "eval_runtime": 626.4542, | |
| "eval_samples_per_second": 101.273, | |
| "eval_steps_per_second": 3.165, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.670843776106934, | |
| "grad_norm": 3.8284225463867188, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.2427, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.670843776106934, | |
| "eval_accuracy": 0.7575149255490681, | |
| "eval_loss": 1.2897659540176392, | |
| "eval_runtime": 624.5044, | |
| "eval_samples_per_second": 101.589, | |
| "eval_steps_per_second": 3.175, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.7230576441102756, | |
| "grad_norm": 3.8263819217681885, | |
| "learning_rate": 8.35e-05, | |
| "loss": 1.2389, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.7230576441102756, | |
| "eval_accuracy": 0.7580776326367493, | |
| "eval_loss": 1.2844797372817993, | |
| "eval_runtime": 627.1556, | |
| "eval_samples_per_second": 101.16, | |
| "eval_steps_per_second": 3.162, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.7752715121136173, | |
| "grad_norm": 3.7820005416870117, | |
| "learning_rate": 8.3e-05, | |
| "loss": 1.2368, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.7752715121136173, | |
| "eval_accuracy": 0.7585746778078989, | |
| "eval_loss": 1.2964411973953247, | |
| "eval_runtime": 635.3019, | |
| "eval_samples_per_second": 99.863, | |
| "eval_steps_per_second": 3.121, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.827485380116959, | |
| "grad_norm": 3.492671012878418, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.2432, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.827485380116959, | |
| "eval_accuracy": 0.7584533967429662, | |
| "eval_loss": NaN, | |
| "eval_runtime": 623.9705, | |
| "eval_samples_per_second": 101.676, | |
| "eval_steps_per_second": 3.178, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "grad_norm": 3.714395761489868, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.2375, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "eval_accuracy": 0.7593634528855384, | |
| "eval_loss": NaN, | |
| "eval_runtime": 623.2531, | |
| "eval_samples_per_second": 101.793, | |
| "eval_steps_per_second": 3.182, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9319131161236425, | |
| "grad_norm": 3.700199604034424, | |
| "learning_rate": 8.15e-05, | |
| "loss": 1.2308, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.9319131161236425, | |
| "eval_accuracy": 0.7597829603112618, | |
| "eval_loss": 1.2819814682006836, | |
| "eval_runtime": 636.6684, | |
| "eval_samples_per_second": 99.648, | |
| "eval_steps_per_second": 3.115, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "grad_norm": 3.390286922454834, | |
| "learning_rate": 8.1e-05, | |
| "loss": 1.2374, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "eval_accuracy": 0.7603004154522052, | |
| "eval_loss": 1.2687220573425293, | |
| "eval_runtime": 630.1502, | |
| "eval_samples_per_second": 100.679, | |
| "eval_steps_per_second": 3.147, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.036340852130326, | |
| "grad_norm": 4.251028060913086, | |
| "learning_rate": 8.05e-05, | |
| "loss": 1.218, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.036340852130326, | |
| "eval_accuracy": 0.760354434691265, | |
| "eval_loss": 1.2784004211425781, | |
| "eval_runtime": 629.6482, | |
| "eval_samples_per_second": 100.759, | |
| "eval_steps_per_second": 3.149, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.0885547201336676, | |
| "grad_norm": 3.714747905731201, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2052, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.0885547201336676, | |
| "eval_accuracy": 0.7609532576337851, | |
| "eval_loss": 1.2718734741210938, | |
| "eval_runtime": 644.9629, | |
| "eval_samples_per_second": 98.367, | |
| "eval_steps_per_second": 3.075, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.1407685881370093, | |
| "grad_norm": 3.934018611907959, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 1.2133, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.1407685881370093, | |
| "eval_accuracy": 0.7613764467527375, | |
| "eval_loss": 1.2647244930267334, | |
| "eval_runtime": 631.2806, | |
| "eval_samples_per_second": 100.499, | |
| "eval_steps_per_second": 3.141, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.192982456140351, | |
| "grad_norm": 3.6540703773498535, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.2072, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.192982456140351, | |
| "eval_accuracy": 0.7618574927985354, | |
| "eval_loss": 1.2702383995056152, | |
| "eval_runtime": 628.7442, | |
| "eval_samples_per_second": 100.904, | |
| "eval_steps_per_second": 3.154, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.2451963241436927, | |
| "grad_norm": 3.0740270614624023, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 1.2051, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.2451963241436927, | |
| "eval_accuracy": 0.7627010896236089, | |
| "eval_loss": 1.272460699081421, | |
| "eval_runtime": 628.6858, | |
| "eval_samples_per_second": 100.914, | |
| "eval_steps_per_second": 3.154, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.2974101921470345, | |
| "grad_norm": 4.153682708740234, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.1978, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.2974101921470345, | |
| "eval_accuracy": 0.7618345683871698, | |
| "eval_loss": 1.2531063556671143, | |
| "eval_runtime": 646.652, | |
| "eval_samples_per_second": 98.11, | |
| "eval_steps_per_second": 3.067, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.3496240601503757, | |
| "grad_norm": 3.2469732761383057, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.2017, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.3496240601503757, | |
| "eval_accuracy": 0.7624854045335798, | |
| "eval_loss": NaN, | |
| "eval_runtime": 639.4743, | |
| "eval_samples_per_second": 99.211, | |
| "eval_steps_per_second": 3.101, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.4018379281537174, | |
| "grad_norm": 3.4860455989837646, | |
| "learning_rate": 7.7e-05, | |
| "loss": 1.2036, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.4018379281537174, | |
| "eval_accuracy": 0.7637592883681051, | |
| "eval_loss": 1.2505515813827515, | |
| "eval_runtime": 626.8199, | |
| "eval_samples_per_second": 101.214, | |
| "eval_steps_per_second": 3.164, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.454051796157059, | |
| "grad_norm": 3.0207018852233887, | |
| "learning_rate": 7.65e-05, | |
| "loss": 1.2039, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.454051796157059, | |
| "eval_accuracy": 0.7636972843433327, | |
| "eval_loss": 1.2614232301712036, | |
| "eval_runtime": 626.1594, | |
| "eval_samples_per_second": 101.321, | |
| "eval_steps_per_second": 3.167, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.506265664160401, | |
| "grad_norm": 2.820202589035034, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.1931, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.506265664160401, | |
| "eval_accuracy": 0.7648100527367832, | |
| "eval_loss": 1.2478718757629395, | |
| "eval_runtime": 654.189, | |
| "eval_samples_per_second": 96.98, | |
| "eval_steps_per_second": 3.031, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.5584795321637426, | |
| "grad_norm": 4.114190101623535, | |
| "learning_rate": 7.55e-05, | |
| "loss": 1.1814, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.5584795321637426, | |
| "eval_accuracy": 0.7652067630736408, | |
| "eval_loss": 1.2430387735366821, | |
| "eval_runtime": 630.6204, | |
| "eval_samples_per_second": 100.604, | |
| "eval_steps_per_second": 3.145, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.6106934001670843, | |
| "grad_norm": 3.205641746520996, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.1934, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.6106934001670843, | |
| "eval_accuracy": 0.7653472241088651, | |
| "eval_loss": 1.2560120820999146, | |
| "eval_runtime": 633.6789, | |
| "eval_samples_per_second": 100.119, | |
| "eval_steps_per_second": 3.129, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.662907268170426, | |
| "grad_norm": 3.286635637283325, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 1.1913, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.662907268170426, | |
| "eval_accuracy": 0.7653709204327143, | |
| "eval_loss": 1.2623989582061768, | |
| "eval_runtime": 636.0478, | |
| "eval_samples_per_second": 99.746, | |
| "eval_steps_per_second": 3.118, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.7151211361737677, | |
| "grad_norm": 3.4579050540924072, | |
| "learning_rate": 7.4e-05, | |
| "loss": 1.1835, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.7151211361737677, | |
| "eval_accuracy": 0.766565753562764, | |
| "eval_loss": 1.2262078523635864, | |
| "eval_runtime": 628.0716, | |
| "eval_samples_per_second": 101.012, | |
| "eval_steps_per_second": 3.157, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.7673350041771094, | |
| "grad_norm": 3.3125433921813965, | |
| "learning_rate": 7.35e-05, | |
| "loss": 1.1821, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.7673350041771094, | |
| "eval_accuracy": 0.7665199419874659, | |
| "eval_loss": 1.249574065208435, | |
| "eval_runtime": 626.5772, | |
| "eval_samples_per_second": 101.253, | |
| "eval_steps_per_second": 3.165, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.819548872180451, | |
| "grad_norm": 3.5980172157287598, | |
| "learning_rate": 7.3e-05, | |
| "loss": 1.1879, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.819548872180451, | |
| "eval_accuracy": 0.7665586654611716, | |
| "eval_loss": 1.2500699758529663, | |
| "eval_runtime": 626.2963, | |
| "eval_samples_per_second": 101.299, | |
| "eval_steps_per_second": 3.166, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.871762740183793, | |
| "grad_norm": 3.683032512664795, | |
| "learning_rate": 7.25e-05, | |
| "loss": 1.1865, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.871762740183793, | |
| "eval_accuracy": 0.767375596206993, | |
| "eval_loss": 1.2403781414031982, | |
| "eval_runtime": 647.9501, | |
| "eval_samples_per_second": 97.913, | |
| "eval_steps_per_second": 3.06, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.9239766081871346, | |
| "grad_norm": 3.2120792865753174, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.1811, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.9239766081871346, | |
| "eval_accuracy": 0.7678975328660842, | |
| "eval_loss": 1.2281544208526611, | |
| "eval_runtime": 641.5519, | |
| "eval_samples_per_second": 98.89, | |
| "eval_steps_per_second": 3.091, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.9761904761904763, | |
| "grad_norm": 3.5708224773406982, | |
| "learning_rate": 7.15e-05, | |
| "loss": 1.1771, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.9761904761904763, | |
| "eval_accuracy": 0.7670722352455037, | |
| "eval_loss": 1.2305808067321777, | |
| "eval_runtime": 629.6736, | |
| "eval_samples_per_second": 100.755, | |
| "eval_steps_per_second": 3.149, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.028404344193818, | |
| "grad_norm": 3.624859094619751, | |
| "learning_rate": 7.1e-05, | |
| "loss": 1.1719, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.028404344193818, | |
| "eval_accuracy": 0.7678675260574238, | |
| "eval_loss": NaN, | |
| "eval_runtime": 653.9665, | |
| "eval_samples_per_second": 97.013, | |
| "eval_steps_per_second": 3.032, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.0806182121971597, | |
| "grad_norm": 3.066574811935425, | |
| "learning_rate": 7.05e-05, | |
| "loss": 1.1625, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.0806182121971597, | |
| "eval_accuracy": 0.7687440934734672, | |
| "eval_loss": 1.2233607769012451, | |
| "eval_runtime": 641.0412, | |
| "eval_samples_per_second": 98.969, | |
| "eval_steps_per_second": 3.093, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.1328320802005014, | |
| "grad_norm": 2.951267719268799, | |
| "learning_rate": 7e-05, | |
| "loss": 1.1667, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.1328320802005014, | |
| "eval_accuracy": 0.7692466135515155, | |
| "eval_loss": 1.2097790241241455, | |
| "eval_runtime": 634.5544, | |
| "eval_samples_per_second": 99.98, | |
| "eval_steps_per_second": 3.125, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.185045948203843, | |
| "grad_norm": 3.134652614593506, | |
| "learning_rate": 6.95e-05, | |
| "loss": 1.161, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.185045948203843, | |
| "eval_accuracy": 0.7697799908068673, | |
| "eval_loss": 1.2196942567825317, | |
| "eval_runtime": 639.2853, | |
| "eval_samples_per_second": 99.241, | |
| "eval_steps_per_second": 3.102, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.2372598162071844, | |
| "grad_norm": 3.199575662612915, | |
| "learning_rate": 6.9e-05, | |
| "loss": 1.1591, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.2372598162071844, | |
| "eval_accuracy": 0.7695065031514984, | |
| "eval_loss": 1.2124276161193848, | |
| "eval_runtime": 682.4327, | |
| "eval_samples_per_second": 92.966, | |
| "eval_steps_per_second": 2.906, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "grad_norm": 2.954310178756714, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 1.1587, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "eval_accuracy": 0.7700210273355362, | |
| "eval_loss": 1.2054550647735596, | |
| "eval_runtime": 639.9415, | |
| "eval_samples_per_second": 99.139, | |
| "eval_steps_per_second": 3.099, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.341687552213868, | |
| "grad_norm": 2.8550124168395996, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.1575, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.341687552213868, | |
| "eval_accuracy": 0.7699819172054071, | |
| "eval_loss": 1.214800238609314, | |
| "eval_runtime": 638.9546, | |
| "eval_samples_per_second": 99.292, | |
| "eval_steps_per_second": 3.104, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.3939014202172095, | |
| "grad_norm": 2.92110013961792, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 1.1523, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.3939014202172095, | |
| "eval_accuracy": 0.769908237207266, | |
| "eval_loss": 1.2140088081359863, | |
| "eval_runtime": 644.5498, | |
| "eval_samples_per_second": 98.43, | |
| "eval_steps_per_second": 3.077, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.4461152882205512, | |
| "grad_norm": 3.346374988555908, | |
| "learning_rate": 6.7e-05, | |
| "loss": 1.1572, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.4461152882205512, | |
| "eval_accuracy": 0.7708846584546473, | |
| "eval_loss": 1.2036925554275513, | |
| "eval_runtime": 640.768, | |
| "eval_samples_per_second": 99.011, | |
| "eval_steps_per_second": 3.095, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.498329156223893, | |
| "grad_norm": 3.251553773880005, | |
| "learning_rate": 6.65e-05, | |
| "loss": 1.1435, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.498329156223893, | |
| "eval_accuracy": 0.7710189865489193, | |
| "eval_loss": 1.2105178833007812, | |
| "eval_runtime": 649.4058, | |
| "eval_samples_per_second": 97.694, | |
| "eval_steps_per_second": 3.054, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.5505430242272347, | |
| "grad_norm": 3.583970785140991, | |
| "learning_rate": 6.6e-05, | |
| "loss": 1.1377, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.5505430242272347, | |
| "eval_accuracy": 0.7721632980503305, | |
| "eval_loss": 1.2023558616638184, | |
| "eval_runtime": 638.8527, | |
| "eval_samples_per_second": 99.308, | |
| "eval_steps_per_second": 3.104, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.6027568922305764, | |
| "grad_norm": 3.588223934173584, | |
| "learning_rate": 6.55e-05, | |
| "loss": 1.1369, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.6027568922305764, | |
| "eval_accuracy": 0.7718440341835643, | |
| "eval_loss": 1.196567416191101, | |
| "eval_runtime": 659.4538, | |
| "eval_samples_per_second": 96.205, | |
| "eval_steps_per_second": 3.007, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.654970760233918, | |
| "grad_norm": 3.4205751419067383, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.1417, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.654970760233918, | |
| "eval_accuracy": 0.7722229829571152, | |
| "eval_loss": 1.1913775205612183, | |
| "eval_runtime": 632.4781, | |
| "eval_samples_per_second": 100.309, | |
| "eval_steps_per_second": 3.135, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.70718462823726, | |
| "grad_norm": 3.300046920776367, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 1.1464, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.70718462823726, | |
| "eval_accuracy": 0.7726826510494151, | |
| "eval_loss": 1.1917306184768677, | |
| "eval_runtime": 632.4563, | |
| "eval_samples_per_second": 100.312, | |
| "eval_steps_per_second": 3.135, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "grad_norm": 3.469228982925415, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.1413, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "eval_accuracy": 0.7728226251433048, | |
| "eval_loss": 1.1853009462356567, | |
| "eval_runtime": 674.7454, | |
| "eval_samples_per_second": 94.025, | |
| "eval_steps_per_second": 2.939, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.8116123642439432, | |
| "grad_norm": 3.0387423038482666, | |
| "learning_rate": 6.35e-05, | |
| "loss": 1.1437, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.8116123642439432, | |
| "eval_accuracy": 0.7730040846904885, | |
| "eval_loss": 1.197380542755127, | |
| "eval_runtime": 642.657, | |
| "eval_samples_per_second": 98.72, | |
| "eval_steps_per_second": 3.086, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.863826232247285, | |
| "grad_norm": 3.7750790119171143, | |
| "learning_rate": 6.3e-05, | |
| "loss": 1.1381, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.863826232247285, | |
| "eval_accuracy": 0.7732420751700277, | |
| "eval_loss": 1.1929783821105957, | |
| "eval_runtime": 635.6748, | |
| "eval_samples_per_second": 99.804, | |
| "eval_steps_per_second": 3.12, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.9160401002506267, | |
| "grad_norm": 2.7454237937927246, | |
| "learning_rate": 6.25e-05, | |
| "loss": 1.1319, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.9160401002506267, | |
| "eval_accuracy": 0.7740359547829445, | |
| "eval_loss": 1.1876792907714844, | |
| "eval_runtime": 646.5493, | |
| "eval_samples_per_second": 98.126, | |
| "eval_steps_per_second": 3.067, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "grad_norm": 3.1448304653167725, | |
| "learning_rate": 6.2e-05, | |
| "loss": 1.1217, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "eval_accuracy": 0.7746280010171029, | |
| "eval_loss": 1.1808606386184692, | |
| "eval_runtime": 651.1984, | |
| "eval_samples_per_second": 97.425, | |
| "eval_steps_per_second": 3.045, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.02046783625731, | |
| "grad_norm": 3.3443033695220947, | |
| "learning_rate": 6.15e-05, | |
| "loss": 1.1297, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.02046783625731, | |
| "eval_accuracy": 0.7744960875837769, | |
| "eval_loss": 1.1759783029556274, | |
| "eval_runtime": 633.4471, | |
| "eval_samples_per_second": 100.155, | |
| "eval_steps_per_second": 3.13, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.072681704260652, | |
| "grad_norm": 3.656834602355957, | |
| "learning_rate": 6.1e-05, | |
| "loss": 1.116, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.072681704260652, | |
| "eval_accuracy": 0.7753487577160052, | |
| "eval_loss": 1.171563744544983, | |
| "eval_runtime": 636.4311, | |
| "eval_samples_per_second": 99.686, | |
| "eval_steps_per_second": 3.116, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.124895572263993, | |
| "grad_norm": 3.2146315574645996, | |
| "learning_rate": 6.05e-05, | |
| "loss": 1.1262, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.124895572263993, | |
| "eval_accuracy": 0.7748815741788464, | |
| "eval_loss": 1.1831566095352173, | |
| "eval_runtime": 626.1408, | |
| "eval_samples_per_second": 101.324, | |
| "eval_steps_per_second": 3.167, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.177109440267335, | |
| "grad_norm": 3.5292935371398926, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1038, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.177109440267335, | |
| "eval_accuracy": 0.775054087211447, | |
| "eval_loss": 1.1849807500839233, | |
| "eval_runtime": 625.7328, | |
| "eval_samples_per_second": 101.39, | |
| "eval_steps_per_second": 3.169, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.2293233082706765, | |
| "grad_norm": 3.17421817779541, | |
| "learning_rate": 5.95e-05, | |
| "loss": 1.1144, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.2293233082706765, | |
| "eval_accuracy": 0.7753948446129992, | |
| "eval_loss": 1.176754117012024, | |
| "eval_runtime": 642.2309, | |
| "eval_samples_per_second": 98.785, | |
| "eval_steps_per_second": 3.088, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.281537176274019, | |
| "grad_norm": 3.185692071914673, | |
| "learning_rate": 5.9e-05, | |
| "loss": 1.1068, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.281537176274019, | |
| "eval_accuracy": 0.7759707359303026, | |
| "eval_loss": 1.1789684295654297, | |
| "eval_runtime": 627.5525, | |
| "eval_samples_per_second": 101.096, | |
| "eval_steps_per_second": 3.16, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.33375104427736, | |
| "grad_norm": 3.033123731613159, | |
| "learning_rate": 5.85e-05, | |
| "loss": 1.1088, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.33375104427736, | |
| "eval_accuracy": 0.7761587182255788, | |
| "eval_loss": 1.1775932312011719, | |
| "eval_runtime": 629.8113, | |
| "eval_samples_per_second": 100.733, | |
| "eval_steps_per_second": 3.149, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "grad_norm": 3.112304210662842, | |
| "learning_rate": 5.8e-05, | |
| "loss": 1.1243, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "eval_accuracy": 0.7764218763015075, | |
| "eval_loss": 1.173019528388977, | |
| "eval_runtime": 655.1515, | |
| "eval_samples_per_second": 96.837, | |
| "eval_steps_per_second": 3.027, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.438178780284043, | |
| "grad_norm": 3.216050624847412, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 1.1196, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.438178780284043, | |
| "eval_accuracy": 0.7775044227327209, | |
| "eval_loss": 1.1682883501052856, | |
| "eval_runtime": 627.0953, | |
| "eval_samples_per_second": 101.17, | |
| "eval_steps_per_second": 3.162, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.4903926482873855, | |
| "grad_norm": 3.4211995601654053, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 1.0985, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.4903926482873855, | |
| "eval_accuracy": 0.7770348312966414, | |
| "eval_loss": 1.1709975004196167, | |
| "eval_runtime": 626.8729, | |
| "eval_samples_per_second": 101.206, | |
| "eval_steps_per_second": 3.163, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.542606516290727, | |
| "grad_norm": 3.0796425342559814, | |
| "learning_rate": 5.65e-05, | |
| "loss": 1.0994, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.542606516290727, | |
| "eval_accuracy": 0.7772564904147641, | |
| "eval_loss": 1.1671358346939087, | |
| "eval_runtime": 626.7232, | |
| "eval_samples_per_second": 101.23, | |
| "eval_steps_per_second": 3.164, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.594820384294069, | |
| "grad_norm": 3.7232813835144043, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.1017, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.594820384294069, | |
| "eval_accuracy": 0.7776795841513819, | |
| "eval_loss": 1.1589510440826416, | |
| "eval_runtime": 639.6154, | |
| "eval_samples_per_second": 99.189, | |
| "eval_steps_per_second": 3.1, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.64703425229741, | |
| "grad_norm": 3.0733375549316406, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 1.1071, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.64703425229741, | |
| "eval_accuracy": 0.7780140462141538, | |
| "eval_loss": 1.1811403036117554, | |
| "eval_runtime": 637.3252, | |
| "eval_samples_per_second": 99.546, | |
| "eval_steps_per_second": 3.111, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.6992481203007515, | |
| "grad_norm": 3.2747066020965576, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 1.0993, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.6992481203007515, | |
| "eval_accuracy": 0.7782215810670943, | |
| "eval_loss": 1.153876781463623, | |
| "eval_runtime": 625.6185, | |
| "eval_samples_per_second": 101.408, | |
| "eval_steps_per_second": 3.17, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.751461988304094, | |
| "grad_norm": 3.6057417392730713, | |
| "learning_rate": 5.45e-05, | |
| "loss": 1.0998, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.751461988304094, | |
| "eval_accuracy": 0.7785066434739322, | |
| "eval_loss": 1.151453971862793, | |
| "eval_runtime": 626.7653, | |
| "eval_samples_per_second": 101.223, | |
| "eval_steps_per_second": 3.164, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.803675856307435, | |
| "grad_norm": 2.923815965652466, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 1.0966, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.803675856307435, | |
| "eval_accuracy": 0.7786641899675826, | |
| "eval_loss": 1.1550520658493042, | |
| "eval_runtime": 626.8592, | |
| "eval_samples_per_second": 101.208, | |
| "eval_steps_per_second": 3.163, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.855889724310777, | |
| "grad_norm": 2.843076229095459, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 1.0977, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.855889724310777, | |
| "eval_accuracy": 0.7794697444838651, | |
| "eval_loss": 1.151402473449707, | |
| "eval_runtime": 625.4097, | |
| "eval_samples_per_second": 101.442, | |
| "eval_steps_per_second": 3.171, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.908103592314118, | |
| "grad_norm": 3.2491214275360107, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 1.0962, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.908103592314118, | |
| "eval_accuracy": 0.7797745180661694, | |
| "eval_loss": 1.1584906578063965, | |
| "eval_runtime": 624.1222, | |
| "eval_samples_per_second": 101.652, | |
| "eval_steps_per_second": 3.177, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.9603174603174605, | |
| "grad_norm": 2.8548121452331543, | |
| "learning_rate": 5.25e-05, | |
| "loss": 1.0933, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.9603174603174605, | |
| "eval_accuracy": 0.7794554886066851, | |
| "eval_loss": 1.1526641845703125, | |
| "eval_runtime": 644.9796, | |
| "eval_samples_per_second": 98.364, | |
| "eval_steps_per_second": 3.075, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 5.012531328320802, | |
| "grad_norm": 3.3320724964141846, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 1.0897, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.012531328320802, | |
| "eval_accuracy": 0.7798054689379954, | |
| "eval_loss": 1.153532862663269, | |
| "eval_runtime": 631.6644, | |
| "eval_samples_per_second": 100.438, | |
| "eval_steps_per_second": 3.139, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.064745196324144, | |
| "grad_norm": 3.1418938636779785, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 1.0727, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.064745196324144, | |
| "eval_accuracy": 0.7807417083407959, | |
| "eval_loss": 1.151798963546753, | |
| "eval_runtime": 628.4763, | |
| "eval_samples_per_second": 100.947, | |
| "eval_steps_per_second": 3.155, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.116959064327485, | |
| "grad_norm": 3.242654800415039, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 1.0846, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.116959064327485, | |
| "eval_accuracy": 0.7803700732682441, | |
| "eval_loss": 1.143381953239441, | |
| "eval_runtime": 626.0556, | |
| "eval_samples_per_second": 101.338, | |
| "eval_steps_per_second": 3.167, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.169172932330827, | |
| "grad_norm": 3.165130615234375, | |
| "learning_rate": 5.05e-05, | |
| "loss": 1.0817, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.169172932330827, | |
| "eval_accuracy": 0.7805262215474418, | |
| "eval_loss": 1.1557620763778687, | |
| "eval_runtime": 638.2573, | |
| "eval_samples_per_second": 99.4, | |
| "eval_steps_per_second": 3.107, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.221386800334169, | |
| "grad_norm": 3.074112892150879, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0725, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.221386800334169, | |
| "eval_accuracy": 0.7803664543536556, | |
| "eval_loss": 1.1497862339019775, | |
| "eval_runtime": 635.6514, | |
| "eval_samples_per_second": 99.808, | |
| "eval_steps_per_second": 3.12, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.273600668337511, | |
| "grad_norm": 3.5100748538970947, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.076, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.273600668337511, | |
| "eval_accuracy": 0.7819452223531255, | |
| "eval_loss": 1.1426007747650146, | |
| "eval_runtime": 627.3013, | |
| "eval_samples_per_second": 101.136, | |
| "eval_steps_per_second": 3.161, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.325814536340852, | |
| "grad_norm": 2.9352290630340576, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.0786, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.325814536340852, | |
| "eval_accuracy": 0.7809871489797384, | |
| "eval_loss": 1.1469109058380127, | |
| "eval_runtime": 651.0651, | |
| "eval_samples_per_second": 97.445, | |
| "eval_steps_per_second": 3.046, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.378028404344194, | |
| "grad_norm": 3.215942859649658, | |
| "learning_rate": 4.85e-05, | |
| "loss": 1.0756, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 5.378028404344194, | |
| "eval_accuracy": 0.7814986150426574, | |
| "eval_loss": 1.1447529792785645, | |
| "eval_runtime": 637.8608, | |
| "eval_samples_per_second": 99.462, | |
| "eval_steps_per_second": 3.109, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 5.430242272347535, | |
| "grad_norm": 3.225491523742676, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0782, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.430242272347535, | |
| "eval_accuracy": 0.781914062234545, | |
| "eval_loss": 1.1327252388000488, | |
| "eval_runtime": 636.3098, | |
| "eval_samples_per_second": 99.705, | |
| "eval_steps_per_second": 3.116, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.482456140350878, | |
| "grad_norm": 3.2323813438415527, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.0718, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 5.482456140350878, | |
| "eval_accuracy": 0.7824720975904875, | |
| "eval_loss": 1.1324148178100586, | |
| "eval_runtime": 634.3817, | |
| "eval_samples_per_second": 100.008, | |
| "eval_steps_per_second": 3.126, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 5.534670008354219, | |
| "grad_norm": 2.8218557834625244, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.0646, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 5.534670008354219, | |
| "eval_accuracy": 0.7827635872108434, | |
| "eval_loss": 1.139945387840271, | |
| "eval_runtime": 685.8615, | |
| "eval_samples_per_second": 92.501, | |
| "eval_steps_per_second": 2.891, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 5.586883876357561, | |
| "grad_norm": 3.848472833633423, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 1.0692, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 5.586883876357561, | |
| "eval_accuracy": 0.7828145845856118, | |
| "eval_loss": 1.1383306980133057, | |
| "eval_runtime": 640.36, | |
| "eval_samples_per_second": 99.074, | |
| "eval_steps_per_second": 3.097, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "grad_norm": 3.0220718383789062, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.0605, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "eval_accuracy": 0.7828910126440776, | |
| "eval_loss": 1.1487065553665161, | |
| "eval_runtime": 639.7494, | |
| "eval_samples_per_second": 99.169, | |
| "eval_steps_per_second": 3.1, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 5.6913116123642435, | |
| "grad_norm": 3.226999044418335, | |
| "learning_rate": 4.55e-05, | |
| "loss": 1.0678, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 5.6913116123642435, | |
| "eval_accuracy": 0.783646746192109, | |
| "eval_loss": 1.1216100454330444, | |
| "eval_runtime": 637.7267, | |
| "eval_samples_per_second": 99.483, | |
| "eval_steps_per_second": 3.109, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 5.743525480367586, | |
| "grad_norm": 3.0716323852539062, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.068, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 5.743525480367586, | |
| "eval_accuracy": 0.7840726797935829, | |
| "eval_loss": 1.1208263635635376, | |
| "eval_runtime": 661.1104, | |
| "eval_samples_per_second": 95.964, | |
| "eval_steps_per_second": 2.999, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 5.795739348370927, | |
| "grad_norm": 3.032036542892456, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 1.0588, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 5.795739348370927, | |
| "eval_accuracy": 0.7842573639982784, | |
| "eval_loss": 1.1288572549819946, | |
| "eval_runtime": 644.8754, | |
| "eval_samples_per_second": 98.38, | |
| "eval_steps_per_second": 3.075, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 5.847953216374269, | |
| "grad_norm": 2.896811008453369, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.0723, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.847953216374269, | |
| "eval_accuracy": 0.7839231768143949, | |
| "eval_loss": 1.11968195438385, | |
| "eval_runtime": 643.1751, | |
| "eval_samples_per_second": 98.64, | |
| "eval_steps_per_second": 3.083, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.90016708437761, | |
| "grad_norm": 3.429722785949707, | |
| "learning_rate": 4.35e-05, | |
| "loss": 1.057, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.90016708437761, | |
| "eval_accuracy": 0.7850116259552448, | |
| "eval_loss": 1.1247141361236572, | |
| "eval_runtime": 639.1011, | |
| "eval_samples_per_second": 99.269, | |
| "eval_steps_per_second": 3.103, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "grad_norm": 2.920243740081787, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.0628, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "eval_accuracy": 0.7846882337268002, | |
| "eval_loss": NaN, | |
| "eval_runtime": 629.1614, | |
| "eval_samples_per_second": 100.837, | |
| "eval_steps_per_second": 3.152, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.004594820384294, | |
| "grad_norm": 2.937222957611084, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.0619, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.004594820384294, | |
| "eval_accuracy": 0.7845797476831528, | |
| "eval_loss": NaN, | |
| "eval_runtime": 628.6009, | |
| "eval_samples_per_second": 100.927, | |
| "eval_steps_per_second": 3.155, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.056808688387636, | |
| "grad_norm": 2.854961395263672, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.0386, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.056808688387636, | |
| "eval_accuracy": 0.7856027081868581, | |
| "eval_loss": 1.1315497159957886, | |
| "eval_runtime": 626.9314, | |
| "eval_samples_per_second": 101.196, | |
| "eval_steps_per_second": 3.163, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.109022556390977, | |
| "grad_norm": 3.4655332565307617, | |
| "learning_rate": 4.15e-05, | |
| "loss": 1.0446, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.109022556390977, | |
| "eval_accuracy": 0.785224256670098, | |
| "eval_loss": NaN, | |
| "eval_runtime": 674.346, | |
| "eval_samples_per_second": 94.081, | |
| "eval_steps_per_second": 2.941, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.161236424394319, | |
| "grad_norm": 2.93487811088562, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.0532, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.161236424394319, | |
| "eval_accuracy": 0.7855367902419893, | |
| "eval_loss": 1.1223821640014648, | |
| "eval_runtime": 630.6681, | |
| "eval_samples_per_second": 100.596, | |
| "eval_steps_per_second": 3.144, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.213450292397661, | |
| "grad_norm": 3.50022554397583, | |
| "learning_rate": 4.05e-05, | |
| "loss": 1.0465, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.213450292397661, | |
| "eval_accuracy": 0.7857459334681388, | |
| "eval_loss": 1.1025224924087524, | |
| "eval_runtime": 625.4125, | |
| "eval_samples_per_second": 101.442, | |
| "eval_steps_per_second": 3.171, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.265664160401003, | |
| "grad_norm": 3.2295687198638916, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0448, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.265664160401003, | |
| "eval_accuracy": 0.7862344900229137, | |
| "eval_loss": 1.104607105255127, | |
| "eval_runtime": 668.9552, | |
| "eval_samples_per_second": 94.839, | |
| "eval_steps_per_second": 2.964, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.317878028404344, | |
| "grad_norm": 3.1386146545410156, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 1.0304, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 6.317878028404344, | |
| "eval_accuracy": 0.7869643708248876, | |
| "eval_loss": 1.1088374853134155, | |
| "eval_runtime": 633.3446, | |
| "eval_samples_per_second": 100.171, | |
| "eval_steps_per_second": 3.131, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 6.370091896407686, | |
| "grad_norm": 3.1814932823181152, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.0354, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 6.370091896407686, | |
| "eval_accuracy": 0.7865247818397344, | |
| "eval_loss": 1.108305811882019, | |
| "eval_runtime": 636.3927, | |
| "eval_samples_per_second": 99.692, | |
| "eval_steps_per_second": 3.116, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 6.4223057644110275, | |
| "grad_norm": 3.155350685119629, | |
| "learning_rate": 3.85e-05, | |
| "loss": 1.0437, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 6.4223057644110275, | |
| "eval_accuracy": 0.7868538533540672, | |
| "eval_loss": NaN, | |
| "eval_runtime": 630.5912, | |
| "eval_samples_per_second": 100.609, | |
| "eval_steps_per_second": 3.145, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 6.474519632414369, | |
| "grad_norm": 2.9944610595703125, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.0347, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 6.474519632414369, | |
| "eval_accuracy": 0.7876532991281543, | |
| "eval_loss": 1.0992841720581055, | |
| "eval_runtime": 649.6208, | |
| "eval_samples_per_second": 97.662, | |
| "eval_steps_per_second": 3.053, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 6.526733500417711, | |
| "grad_norm": 2.89630126953125, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.041, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 6.526733500417711, | |
| "eval_accuracy": 0.7875077762136071, | |
| "eval_loss": 1.1099497079849243, | |
| "eval_runtime": 631.7247, | |
| "eval_samples_per_second": 100.428, | |
| "eval_steps_per_second": 3.139, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "grad_norm": 3.2351603507995605, | |
| "learning_rate": 3.7e-05, | |
| "loss": 1.0388, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "eval_accuracy": 0.78800361330496, | |
| "eval_loss": 1.11760413646698, | |
| "eval_runtime": 629.3158, | |
| "eval_samples_per_second": 100.813, | |
| "eval_steps_per_second": 3.151, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 6.631161236424394, | |
| "grad_norm": 3.2929039001464844, | |
| "learning_rate": 3.65e-05, | |
| "loss": 1.0383, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 6.631161236424394, | |
| "eval_accuracy": 0.7879278944035151, | |
| "eval_loss": 1.1053088903427124, | |
| "eval_runtime": 627.1228, | |
| "eval_samples_per_second": 101.165, | |
| "eval_steps_per_second": 3.162, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 6.683375104427736, | |
| "grad_norm": 2.9945790767669678, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.0373, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 6.683375104427736, | |
| "eval_accuracy": 0.7881556360589377, | |
| "eval_loss": 1.1067023277282715, | |
| "eval_runtime": 664.2963, | |
| "eval_samples_per_second": 95.504, | |
| "eval_steps_per_second": 2.985, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 6.735588972431078, | |
| "grad_norm": 3.4380545616149902, | |
| "learning_rate": 3.55e-05, | |
| "loss": 1.0382, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 6.735588972431078, | |
| "eval_accuracy": 0.7882482565200516, | |
| "eval_loss": 1.116519570350647, | |
| "eval_runtime": 629.5022, | |
| "eval_samples_per_second": 100.783, | |
| "eval_steps_per_second": 3.15, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 6.787802840434419, | |
| "grad_norm": 3.368530750274658, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0277, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 6.787802840434419, | |
| "eval_accuracy": 0.7891739950636399, | |
| "eval_loss": 1.0915908813476562, | |
| "eval_runtime": 632.7041, | |
| "eval_samples_per_second": 100.273, | |
| "eval_steps_per_second": 3.134, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 6.840016708437761, | |
| "grad_norm": 3.3649165630340576, | |
| "learning_rate": 3.45e-05, | |
| "loss": 1.0337, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 6.840016708437761, | |
| "eval_accuracy": 0.7883815172332914, | |
| "eval_loss": 1.1139589548110962, | |
| "eval_runtime": 640.7807, | |
| "eval_samples_per_second": 99.009, | |
| "eval_steps_per_second": 3.095, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 6.8922305764411025, | |
| "grad_norm": 3.242825746536255, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.033, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.8922305764411025, | |
| "eval_accuracy": 0.789335987558676, | |
| "eval_loss": 1.0902156829833984, | |
| "eval_runtime": 634.5325, | |
| "eval_samples_per_second": 99.984, | |
| "eval_steps_per_second": 3.125, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 2.7917425632476807, | |
| "learning_rate": 3.35e-05, | |
| "loss": 1.0298, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "eval_accuracy": 0.7897472088960492, | |
| "eval_loss": 1.0988086462020874, | |
| "eval_runtime": 636.5285, | |
| "eval_samples_per_second": 99.67, | |
| "eval_steps_per_second": 3.115, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 6.996658312447786, | |
| "grad_norm": 2.9491727352142334, | |
| "learning_rate": 3.3e-05, | |
| "loss": 1.0325, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 6.996658312447786, | |
| "eval_accuracy": 0.7894358436331047, | |
| "eval_loss": 1.097759485244751, | |
| "eval_runtime": 631.2152, | |
| "eval_samples_per_second": 100.509, | |
| "eval_steps_per_second": 3.142, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 7.048872180451128, | |
| "grad_norm": 3.81508469581604, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 1.013, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.048872180451128, | |
| "eval_accuracy": 0.7897459597854384, | |
| "eval_loss": 1.1025677919387817, | |
| "eval_runtime": 667.978, | |
| "eval_samples_per_second": 94.978, | |
| "eval_steps_per_second": 2.969, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.101086048454469, | |
| "grad_norm": 2.6366662979125977, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.0199, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.101086048454469, | |
| "eval_accuracy": 0.7899999141985645, | |
| "eval_loss": 1.1025761365890503, | |
| "eval_runtime": 630.4778, | |
| "eval_samples_per_second": 100.627, | |
| "eval_steps_per_second": 3.145, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.1532999164578115, | |
| "grad_norm": 3.155102014541626, | |
| "learning_rate": 3.15e-05, | |
| "loss": 1.0157, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 7.1532999164578115, | |
| "eval_accuracy": 0.7898970170010094, | |
| "eval_loss": 1.1064562797546387, | |
| "eval_runtime": 625.4866, | |
| "eval_samples_per_second": 101.43, | |
| "eval_steps_per_second": 3.17, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 7.205513784461153, | |
| "grad_norm": 2.9615111351013184, | |
| "learning_rate": 3.1e-05, | |
| "loss": 1.0137, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 7.205513784461153, | |
| "eval_accuracy": 0.7904352669747394, | |
| "eval_loss": 1.0876559019088745, | |
| "eval_runtime": 646.3487, | |
| "eval_samples_per_second": 98.156, | |
| "eval_steps_per_second": 3.068, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 7.257727652464495, | |
| "grad_norm": 3.723841905593872, | |
| "learning_rate": 3.05e-05, | |
| "loss": 1.0137, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 7.257727652464495, | |
| "eval_accuracy": 0.7899743837210673, | |
| "eval_loss": 1.0912542343139648, | |
| "eval_runtime": 652.0884, | |
| "eval_samples_per_second": 97.292, | |
| "eval_steps_per_second": 3.041, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 7.309941520467836, | |
| "grad_norm": 3.6348705291748047, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0167, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 7.309941520467836, | |
| "eval_accuracy": 0.7903721199263885, | |
| "eval_loss": NaN, | |
| "eval_runtime": 630.9965, | |
| "eval_samples_per_second": 100.544, | |
| "eval_steps_per_second": 3.143, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 7.362155388471178, | |
| "grad_norm": 3.2423222064971924, | |
| "learning_rate": 2.95e-05, | |
| "loss": 1.0089, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 7.362155388471178, | |
| "eval_accuracy": 0.7906133078105837, | |
| "eval_loss": 1.0956330299377441, | |
| "eval_runtime": 633.5776, | |
| "eval_samples_per_second": 100.135, | |
| "eval_steps_per_second": 3.13, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 7.41436925647452, | |
| "grad_norm": 3.15704083442688, | |
| "learning_rate": 2.9e-05, | |
| "loss": 1.0055, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 7.41436925647452, | |
| "eval_accuracy": 0.7912404700602101, | |
| "eval_loss": NaN, | |
| "eval_runtime": 649.7229, | |
| "eval_samples_per_second": 97.646, | |
| "eval_steps_per_second": 3.052, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 7.466583124477861, | |
| "grad_norm": 3.387059211730957, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 1.0135, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 7.466583124477861, | |
| "eval_accuracy": 0.7914904478088925, | |
| "eval_loss": 1.0922181606292725, | |
| "eval_runtime": 633.8807, | |
| "eval_samples_per_second": 100.087, | |
| "eval_steps_per_second": 3.128, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "grad_norm": 3.3285696506500244, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.0041, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "eval_accuracy": 0.7919370824790591, | |
| "eval_loss": 1.0766669511795044, | |
| "eval_runtime": 640.2628, | |
| "eval_samples_per_second": 99.089, | |
| "eval_steps_per_second": 3.097, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 7.571010860484545, | |
| "grad_norm": 3.0725443363189697, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.9982, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 7.571010860484545, | |
| "eval_accuracy": 0.7914295287916221, | |
| "eval_loss": 1.0892815589904785, | |
| "eval_runtime": 636.4517, | |
| "eval_samples_per_second": 99.682, | |
| "eval_steps_per_second": 3.116, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 7.6232247284878865, | |
| "grad_norm": 3.2176389694213867, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 1.0066, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 7.6232247284878865, | |
| "eval_accuracy": 0.7917736483647045, | |
| "eval_loss": 1.092371940612793, | |
| "eval_runtime": 621.243, | |
| "eval_samples_per_second": 102.123, | |
| "eval_steps_per_second": 3.192, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 7.675438596491228, | |
| "grad_norm": 2.7997543811798096, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 1.0073, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 7.675438596491228, | |
| "eval_accuracy": 0.7922002806270909, | |
| "eval_loss": 1.0904649496078491, | |
| "eval_runtime": 620.7253, | |
| "eval_samples_per_second": 102.208, | |
| "eval_steps_per_second": 3.195, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 7.72765246449457, | |
| "grad_norm": 3.626774549484253, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.0015, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 7.72765246449457, | |
| "eval_accuracy": 0.7917596219688585, | |
| "eval_loss": 1.0867533683776855, | |
| "eval_runtime": 618.3607, | |
| "eval_samples_per_second": 102.599, | |
| "eval_steps_per_second": 3.207, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 7.779866332497911, | |
| "grad_norm": 2.8419744968414307, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 1.0049, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 7.779866332497911, | |
| "eval_accuracy": 0.7928055223543873, | |
| "eval_loss": 1.0817667245864868, | |
| "eval_runtime": 617.1968, | |
| "eval_samples_per_second": 102.792, | |
| "eval_steps_per_second": 3.213, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 7.832080200501253, | |
| "grad_norm": 2.822751760482788, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.9957, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 7.832080200501253, | |
| "eval_accuracy": 0.7932603950321594, | |
| "eval_loss": 1.0652581453323364, | |
| "eval_runtime": 616.9484, | |
| "eval_samples_per_second": 102.834, | |
| "eval_steps_per_second": 3.214, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 7.884294068504595, | |
| "grad_norm": 2.9094507694244385, | |
| "learning_rate": 2.45e-05, | |
| "loss": 1.0042, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 7.884294068504595, | |
| "eval_accuracy": 0.7934020652494004, | |
| "eval_loss": 1.0694156885147095, | |
| "eval_runtime": 617.9112, | |
| "eval_samples_per_second": 102.673, | |
| "eval_steps_per_second": 3.209, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "grad_norm": 2.5249078273773193, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.0031, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "eval_accuracy": 0.7935896894661485, | |
| "eval_loss": 1.0808907747268677, | |
| "eval_runtime": 618.1232, | |
| "eval_samples_per_second": 102.638, | |
| "eval_steps_per_second": 3.208, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 7.988721804511278, | |
| "grad_norm": 3.253228187561035, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.9979, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 7.988721804511278, | |
| "eval_accuracy": 0.7933455565489351, | |
| "eval_loss": 1.0679088830947876, | |
| "eval_runtime": 618.0488, | |
| "eval_samples_per_second": 102.65, | |
| "eval_steps_per_second": 3.208, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 8.04093567251462, | |
| "grad_norm": 2.6395580768585205, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.9825, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 8.04093567251462, | |
| "eval_accuracy": 0.7935253800154058, | |
| "eval_loss": 1.07496976852417, | |
| "eval_runtime": 617.9799, | |
| "eval_samples_per_second": 102.662, | |
| "eval_steps_per_second": 3.209, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 8.093149540517961, | |
| "grad_norm": 3.0883655548095703, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.9926, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 8.093149540517961, | |
| "eval_accuracy": 0.7942360815200339, | |
| "eval_loss": 1.0814706087112427, | |
| "eval_runtime": 618.6019, | |
| "eval_samples_per_second": 102.559, | |
| "eval_steps_per_second": 3.206, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 8.145363408521304, | |
| "grad_norm": 3.356086015701294, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.9897, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 8.145363408521304, | |
| "eval_accuracy": 0.7933161062503574, | |
| "eval_loss": NaN, | |
| "eval_runtime": 617.0688, | |
| "eval_samples_per_second": 102.813, | |
| "eval_steps_per_second": 3.214, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 8.197577276524646, | |
| "grad_norm": 3.31772780418396, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.9871, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 8.197577276524646, | |
| "eval_accuracy": 0.7943132034313805, | |
| "eval_loss": 1.0676552057266235, | |
| "eval_runtime": 623.639, | |
| "eval_samples_per_second": 101.73, | |
| "eval_steps_per_second": 3.18, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 8.249791144527986, | |
| "grad_norm": 3.543698310852051, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.0001, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 8.249791144527986, | |
| "eval_accuracy": 0.7938355754922207, | |
| "eval_loss": 1.0734678506851196, | |
| "eval_runtime": 617.8987, | |
| "eval_samples_per_second": 102.675, | |
| "eval_steps_per_second": 3.209, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 8.302005012531328, | |
| "grad_norm": 2.846734046936035, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.987, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 8.302005012531328, | |
| "eval_accuracy": 0.7944878324075739, | |
| "eval_loss": 1.0765976905822754, | |
| "eval_runtime": 618.672, | |
| "eval_samples_per_second": 102.547, | |
| "eval_steps_per_second": 3.205, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 8.35421888053467, | |
| "grad_norm": 2.871553897857666, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9806, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 8.35421888053467, | |
| "eval_accuracy": 0.794915350004184, | |
| "eval_loss": 1.0597549676895142, | |
| "eval_runtime": 618.1763, | |
| "eval_samples_per_second": 102.629, | |
| "eval_steps_per_second": 3.208, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 8.406432748538013, | |
| "grad_norm": 2.959845542907715, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.984, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 8.406432748538013, | |
| "eval_accuracy": 0.7946179832675856, | |
| "eval_loss": 1.0595872402191162, | |
| "eval_runtime": 617.7039, | |
| "eval_samples_per_second": 102.708, | |
| "eval_steps_per_second": 3.21, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 8.458646616541353, | |
| "grad_norm": 2.7344448566436768, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.9822, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 8.458646616541353, | |
| "eval_accuracy": 0.7951521505243792, | |
| "eval_loss": 1.0736610889434814, | |
| "eval_runtime": 618.3207, | |
| "eval_samples_per_second": 102.605, | |
| "eval_steps_per_second": 3.207, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 8.510860484544695, | |
| "grad_norm": 3.7596890926361084, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.9791, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 8.510860484544695, | |
| "eval_accuracy": 0.795253175296364, | |
| "eval_loss": 1.0666213035583496, | |
| "eval_runtime": 620.245, | |
| "eval_samples_per_second": 102.287, | |
| "eval_steps_per_second": 3.197, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 8.563074352548037, | |
| "grad_norm": 2.727389335632324, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.9804, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 8.563074352548037, | |
| "eval_accuracy": 0.7953146030482228, | |
| "eval_loss": 1.0600041151046753, | |
| "eval_runtime": 618.8007, | |
| "eval_samples_per_second": 102.526, | |
| "eval_steps_per_second": 3.205, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 8.615288220551378, | |
| "grad_norm": 2.9671452045440674, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.9703, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 8.615288220551378, | |
| "eval_accuracy": 0.7963827692254926, | |
| "eval_loss": 1.0601171255111694, | |
| "eval_runtime": 616.1581, | |
| "eval_samples_per_second": 102.965, | |
| "eval_steps_per_second": 3.218, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 8.66750208855472, | |
| "grad_norm": 2.9628329277038574, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.9768, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 8.66750208855472, | |
| "eval_accuracy": 0.7952524450397661, | |
| "eval_loss": 1.0768334865570068, | |
| "eval_runtime": 616.7531, | |
| "eval_samples_per_second": 102.866, | |
| "eval_steps_per_second": 3.215, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 8.719715956558062, | |
| "grad_norm": 3.6838812828063965, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.9722, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 8.719715956558062, | |
| "eval_accuracy": 0.796164760278738, | |
| "eval_loss": NaN, | |
| "eval_runtime": 616.0678, | |
| "eval_samples_per_second": 102.981, | |
| "eval_steps_per_second": 3.219, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 3.7852931022644043, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.9813, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "eval_accuracy": 0.7964795153494032, | |
| "eval_loss": NaN, | |
| "eval_runtime": 616.9446, | |
| "eval_samples_per_second": 102.834, | |
| "eval_steps_per_second": 3.214, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 8.824143692564745, | |
| "grad_norm": 2.85229229927063, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.9732, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 8.824143692564745, | |
| "eval_accuracy": 0.7960116027401912, | |
| "eval_loss": 1.061353325843811, | |
| "eval_runtime": 624.2524, | |
| "eval_samples_per_second": 101.63, | |
| "eval_steps_per_second": 3.177, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 8.876357560568087, | |
| "grad_norm": 3.463848114013672, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9764, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 8.876357560568087, | |
| "eval_accuracy": 0.7963621203122637, | |
| "eval_loss": 1.049773097038269, | |
| "eval_runtime": 618.458, | |
| "eval_samples_per_second": 102.583, | |
| "eval_steps_per_second": 3.206, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "grad_norm": 2.7122750282287598, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.9829, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "eval_accuracy": 0.7965627753528177, | |
| "eval_loss": 1.069191813468933, | |
| "eval_runtime": 617.5416, | |
| "eval_samples_per_second": 102.735, | |
| "eval_steps_per_second": 3.211, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 8.980785296574771, | |
| "grad_norm": 2.976637840270996, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.9741, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 8.980785296574771, | |
| "eval_accuracy": 0.7967220469069121, | |
| "eval_loss": 1.055444598197937, | |
| "eval_runtime": 614.0894, | |
| "eval_samples_per_second": 103.312, | |
| "eval_steps_per_second": 3.229, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 9.032999164578111, | |
| "grad_norm": 2.89384126663208, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.9648, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 9.032999164578111, | |
| "eval_accuracy": 0.797160834758222, | |
| "eval_loss": 1.058738350868225, | |
| "eval_runtime": 615.784, | |
| "eval_samples_per_second": 103.028, | |
| "eval_steps_per_second": 3.22, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 9.085213032581454, | |
| "grad_norm": 3.009979248046875, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.9606, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 9.085213032581454, | |
| "eval_accuracy": 0.7971404883516657, | |
| "eval_loss": 1.0621484518051147, | |
| "eval_runtime": 618.535, | |
| "eval_samples_per_second": 102.57, | |
| "eval_steps_per_second": 3.206, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 9.137426900584796, | |
| "grad_norm": 3.301671266555786, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.9682, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 9.137426900584796, | |
| "eval_accuracy": 0.7973926994154387, | |
| "eval_loss": 1.0569199323654175, | |
| "eval_runtime": 616.4943, | |
| "eval_samples_per_second": 102.909, | |
| "eval_steps_per_second": 3.217, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 9.189640768588138, | |
| "grad_norm": 2.932385206222534, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9644, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 9.189640768588138, | |
| "eval_accuracy": 0.7974666942085554, | |
| "eval_loss": 1.0533905029296875, | |
| "eval_runtime": 620.5893, | |
| "eval_samples_per_second": 102.23, | |
| "eval_steps_per_second": 3.195, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 9.241854636591478, | |
| "grad_norm": 3.1355013847351074, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.9658, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 9.241854636591478, | |
| "eval_accuracy": 0.7977740520228777, | |
| "eval_loss": 1.0476280450820923, | |
| "eval_runtime": 618.7726, | |
| "eval_samples_per_second": 102.53, | |
| "eval_steps_per_second": 3.205, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 9.29406850459482, | |
| "grad_norm": 3.353086233139038, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.9641, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 9.29406850459482, | |
| "eval_accuracy": 0.7973864228195986, | |
| "eval_loss": 1.041751742362976, | |
| "eval_runtime": 614.3896, | |
| "eval_samples_per_second": 103.262, | |
| "eval_steps_per_second": 3.228, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 9.346282372598163, | |
| "grad_norm": 2.8681247234344482, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.963, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 9.346282372598163, | |
| "eval_accuracy": 0.798292731668655, | |
| "eval_loss": 1.0465270280838013, | |
| "eval_runtime": 615.1098, | |
| "eval_samples_per_second": 103.141, | |
| "eval_steps_per_second": 3.224, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 9.398496240601503, | |
| "grad_norm": 3.2927052974700928, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9562, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 9.398496240601503, | |
| "eval_accuracy": 0.798140924991145, | |
| "eval_loss": 1.0476934909820557, | |
| "eval_runtime": 616.723, | |
| "eval_samples_per_second": 102.871, | |
| "eval_steps_per_second": 3.215, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 9.450710108604845, | |
| "grad_norm": 3.1475863456726074, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.9605, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 9.450710108604845, | |
| "eval_accuracy": 0.7977010212122122, | |
| "eval_loss": 1.0535070896148682, | |
| "eval_runtime": 615.5187, | |
| "eval_samples_per_second": 103.072, | |
| "eval_steps_per_second": 3.222, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 9.502923976608187, | |
| "grad_norm": 3.662548780441284, | |
| "learning_rate": 9e-06, | |
| "loss": 0.9692, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 9.502923976608187, | |
| "eval_accuracy": 0.7981552799718717, | |
| "eval_loss": 1.0515964031219482, | |
| "eval_runtime": 615.1273, | |
| "eval_samples_per_second": 103.138, | |
| "eval_steps_per_second": 3.224, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 9.55513784461153, | |
| "grad_norm": 3.5501346588134766, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.966, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 9.55513784461153, | |
| "eval_accuracy": 0.798176074856451, | |
| "eval_loss": 1.0500941276550293, | |
| "eval_runtime": 616.9498, | |
| "eval_samples_per_second": 102.833, | |
| "eval_steps_per_second": 3.214, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 9.60735171261487, | |
| "grad_norm": 3.102790117263794, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9614, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 9.60735171261487, | |
| "eval_accuracy": 0.7986090909415797, | |
| "eval_loss": 1.0444718599319458, | |
| "eval_runtime": 614.407, | |
| "eval_samples_per_second": 103.259, | |
| "eval_steps_per_second": 3.228, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 9.659565580618212, | |
| "grad_norm": 3.504926919937134, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.9606, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 9.659565580618212, | |
| "eval_accuracy": 0.7988812850380725, | |
| "eval_loss": 1.0497493743896484, | |
| "eval_runtime": 615.4873, | |
| "eval_samples_per_second": 103.078, | |
| "eval_steps_per_second": 3.222, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 9.711779448621554, | |
| "grad_norm": 3.648078203201294, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.9507, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 9.711779448621554, | |
| "eval_accuracy": 0.798754673907639, | |
| "eval_loss": 1.0451751947402954, | |
| "eval_runtime": 615.9945, | |
| "eval_samples_per_second": 102.993, | |
| "eval_steps_per_second": 3.219, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 9.763993316624896, | |
| "grad_norm": 3.316922903060913, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.9584, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 9.763993316624896, | |
| "eval_accuracy": 0.7987269314893306, | |
| "eval_loss": 1.0408653020858765, | |
| "eval_runtime": 618.0487, | |
| "eval_samples_per_second": 102.65, | |
| "eval_steps_per_second": 3.208, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 9.816207184628237, | |
| "grad_norm": 2.943169116973877, | |
| "learning_rate": 6e-06, | |
| "loss": 0.9572, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 9.816207184628237, | |
| "eval_accuracy": 0.7989658711747047, | |
| "eval_loss": 1.0431112051010132, | |
| "eval_runtime": 619.8767, | |
| "eval_samples_per_second": 102.348, | |
| "eval_steps_per_second": 3.199, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 9.868421052631579, | |
| "grad_norm": 2.713733196258545, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.9552, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 9.868421052631579, | |
| "eval_accuracy": 0.7991049545929755, | |
| "eval_loss": 1.0465214252471924, | |
| "eval_runtime": 615.6457, | |
| "eval_samples_per_second": 103.051, | |
| "eval_steps_per_second": 3.221, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 9.920634920634921, | |
| "grad_norm": 3.5131709575653076, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9542, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 9.920634920634921, | |
| "eval_accuracy": 0.7993720778709359, | |
| "eval_loss": 1.043885350227356, | |
| "eval_runtime": 616.5178, | |
| "eval_samples_per_second": 102.905, | |
| "eval_steps_per_second": 3.216, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 9.972848788638263, | |
| "grad_norm": 3.0255749225616455, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.9505, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 9.972848788638263, | |
| "eval_accuracy": 0.7994996190810016, | |
| "eval_loss": 1.0470497608184814, | |
| "eval_runtime": 616.7354, | |
| "eval_samples_per_second": 102.869, | |
| "eval_steps_per_second": 3.215, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 10.025062656641603, | |
| "grad_norm": 4.134832382202148, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9472, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 10.025062656641603, | |
| "eval_accuracy": 0.7994244357414679, | |
| "eval_loss": 1.0445414781570435, | |
| "eval_runtime": 617.5631, | |
| "eval_samples_per_second": 102.731, | |
| "eval_steps_per_second": 3.211, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 10.077276524644946, | |
| "grad_norm": 3.3011510372161865, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.9467, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 10.077276524644946, | |
| "eval_accuracy": 0.7997884398890616, | |
| "eval_loss": 1.0422955751419067, | |
| "eval_runtime": 617.5269, | |
| "eval_samples_per_second": 102.737, | |
| "eval_steps_per_second": 3.211, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 10.129490392648288, | |
| "grad_norm": 3.65120792388916, | |
| "learning_rate": 3e-06, | |
| "loss": 0.9497, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 10.129490392648288, | |
| "eval_accuracy": 0.7994027266856298, | |
| "eval_loss": 1.0433976650238037, | |
| "eval_runtime": 615.8756, | |
| "eval_samples_per_second": 103.013, | |
| "eval_steps_per_second": 3.22, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 10.18170426065163, | |
| "grad_norm": 3.1329991817474365, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.9575, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 10.18170426065163, | |
| "eval_accuracy": 0.8000353846923677, | |
| "eval_loss": 1.0379133224487305, | |
| "eval_runtime": 615.8872, | |
| "eval_samples_per_second": 103.011, | |
| "eval_steps_per_second": 3.22, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 10.23391812865497, | |
| "grad_norm": 3.3150088787078857, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.9478, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 10.23391812865497, | |
| "eval_accuracy": 0.7997029578657322, | |
| "eval_loss": 1.039129614830017, | |
| "eval_runtime": 615.7144, | |
| "eval_samples_per_second": 103.04, | |
| "eval_steps_per_second": 3.221, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 10.286131996658312, | |
| "grad_norm": 2.9896209239959717, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.9428, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 10.286131996658312, | |
| "eval_accuracy": 0.799764992788252, | |
| "eval_loss": 1.0273813009262085, | |
| "eval_runtime": 615.1763, | |
| "eval_samples_per_second": 103.13, | |
| "eval_steps_per_second": 3.223, | |
| "step": 98500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "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": 8.317352524927468e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |